Problems of Shapour River Salinity Rising Over Recent Prolonged Streamow Reduction Period and Solutions of River Salinity Management: An Originally Freshwater River Intensively Salinized by Natural Salinity Sources

The Shapour river with catchment area of 4254 km 2 is a major river system in southern Iran. While the upstream river ow (the upper Shapour river) is fresh, it becomes extremely salinized at the downstream conuence of Shekastian salty tributary and the entering nearby Boushigan brine spring. The river then passes through the Khesht plain and nally discharges into the Raeisali-Delvari storage dam, which went into operation in 2009. Over the 2006–2019 period, reduced precipitation and over-utilization of freshwater resources resulted in ~ 72% streamow reduction in the Shapour river. Consequently, the ratios of unused salty/brine water of Shekastian tributary and Boushigan spring to fresh-outow of the upper Shapour river increased by ~ 3 times and river salinity uctuation domain at the Khesht plain inlet dramatically increased from 2.1-4.0 dS m − 1 to 3.7–26.0 dS m − 1 . It also resulted in disappearance of most river aquatic species and caused major economic damages in the middle Shapour river. On the seasonal time-scale, consecutive processes of salt accumulation during irrigation season of the Khesht plain date orchards and then salt drainage during rainy season have adjusted salinity uctuation domain from 3.7–26.0 dS m − 1 at the plain inlet to 5.2–8.9 dS m − 1 at the plain outlet. In the lower Shapour river, storage/mixing of fresh/salty inow waters in the Raeisali-Delvari reservoir has adjusted strong salinity uctuation domain from 0.9–10.7 dS m − 1 at the reservoir inlet to 3.6–5.5 dS m − 1 at the reservoir outlet. the Shapour river, by on of with natural saline/fresh controlled


Introduction
In addition to water scarcity, salinity of water resources is a major problem in many parts of the world (Hart et al. 1990). Salinity causes yield reduction of agricultural crops, limitation in diversity of crop species, corrosion of water conveyance systems, limitation in biodiversity of both aquatic and riparian species, and increase in costs of drinking water treatment (Miyamoto, 2007 For example, salinization of the Ganges river in Bangladesh has resulted in losses of millions of dollars related to crop and industrial machinery damage and has imposed major risks to human health (Mirza, 1998). Economic models prepared by United States Bureau of Reclamation suggested that salinity contributes to more than 300 million dollars per year in economic damage across the Colorado river basin, with roughly half occurring in the agricultural sector (Borda, 2004).
Salinization of rivers might be ampli ed by droughts, periods of less than normal precipitation, stream ow reduction, and damming ( southern USA increases compared to non-drought periods (Jones and van Vliet, 2018). In the Colorado river catchment, periods of greater than normal precipitation corresponded with decreasing river salinity and periods of less than normal precipitation corresponded with increasing river salinity (Tillman et al. 2019). In hot and dry regions, evaporation from reservoir water surface area can increase river salinity at downstream of dam. For example, due to evaporation, out ow salinity of the Amistad International Reservoir, located at the Texas-Mexico border, has been averaging 10% larger than the in ow salinity . Comparison between measured long-term salinity data in the monitoring stations While several studies have addressed the issue of increased river salinity at downstream of dams, it is possible that reservoir of a storage dam could reduce the river salinity uctuations, as in ow mixes with stored water in dam reservoir (Inosako et al. 2006). As a matter of fact, it appears that investigation of reservoirs as regulators of river ow salinity has not been the focus of previous studies.
Most of the world's subaerial salt exposures are located in the arid and semiarid Zagros Mountains, Iran (Kent, 1979;Bosak et al. 1998 This research aims to investigate the in uence of natural salinity sources, over-utilization of freshwater resources, agricultural drainage and damming on Shapour river salinity during the recent reduced stream ow period (2006-2019). In this respect, the role of the Raeisali-Delvari storage dam is investigated as a water quality regulator, intended to improve the quality of the reservoir out ow water.
The main focus of this research will be on detailed data analysis, leading to solutions for salinity reduction/adjustment of in uenced rivers by natural salinity sources.

Situation And Hydrogeology Of Study Area
The Shapour river catchment with surface area of 4254 Km 2 , located in a semi-arid climate in the southern parts of the Zagros mountains, southern Iran, is the study area. Situation and geology map of the Shapour river catchment and location of the measurement stations are shown in Fig. 1.
The 220 Km long Shapour river originates from the Sasan spring at the ancient place of Tnageh- The geological formations in the Shapour river catchment in ascending stratigraphic order are: Hormuz salt, Ilam-Sarvak limestone, Pabdeh-Gourpi marl, shale and limestone, Asmari limestone, Gachsaran salt, gypsum, and marl, Mishan marl, shale and limestone, Aghajari sandstone and marl, Lahbari ood-plain deposits, Bakhtiari conglomerate, and Quaternary alluvial deposits. Among these formations, Ilam-Sarvak, Asmari, Bakhtiari, and Quaternary deposits are important from the perspective of development of fresh groundwater resources, mainly located on east side of river catchment. The outcrop of the Gachsaran formation is the most extended formation in the study area and along with two diapirs of Hormuz salt formation are the main natural salinity sources for quality degradation of the Shapour river.
There are 6 discharge and salinity measurement stations in the study area, identi ed by symbols S1 to S6 on Fig. 1. Except for station S2, other stations are located on the Shapour river. Stations S1 (outlet of the upper Shapour), S2 (outlet of the Shekastian salty tributary), S4 (outlet of the Khesht plain), and S6 (outlet of dam reservoir) are hydrometric stations. In stations S3 (inlet of the Kheshat plain) and S5 (inlet of dam reservoir), discharge and salinity of the Shapour river has monthly been measured for some years.
In the study area, Shekastian salty tributary, Boushigan brine spring, and Malicheh brackish tributary are the permanent natural salinity sources for the Shapour river. Surface runoff of Gachsaran and Hormoz formations are non-permanent salinity sources for the Shapour river. Additionally, evaporation from stored water in the Raeisali-Delvari reservoir is a salinity source for the Shapour river below the dam.
Discharge and salinity of Shekastian and Malicheh tributaries are highly variable and dependent on seasonal and annual variability in precipitation. In contrast, discharge and salinity of Boushigan brine spring are approximately constant values and are not in uenced by seasonal or annual precipitation variability. Based on eld measurements, discharge and salt concentration values of the Boushigan brine spring are respectively ~ 11 l s − 1 and ~ 250 g s − 1 .
The oldest rain gauge station in the study area was installed in 1971 at a close distance to the con uence of Shekastian salty tributary with the Shapour river. Using measured precipitation data in this rain gauge station, variation in annual precipitation in the Shapour river catchment is plotted for the 1971-2019 period (Fig. 2). As shown, the average annual precipitation is 452 mm. Additionally

Results And Discussions
To identify time range and intensity of the Shapour river stream ow reduction period, variation in river annual out ow volume at hydrometric stations of S1 (outlet of the upper Shapour river), S4 (middle of middle Shapour river), and S6 (inlet of lower Shapour river) are plotted in Fig. 3. As it can be seen, an unprecedented common stream ow reduction period in the three hydrometric stations was initiated in 2006 and is to be continued. Based on this common variation, average of annual river out ow volume at hydrometric stations were separately calculated for two different time ranges before and after 2006. The results were then plotted in Fig. 3 as the average line. Compared to the period prior to 2006, average annual out ow volume of the Shapour river at stations S1, S4 and S6 decreased by 69, 74, and 74%, respectively. Overall, stream ow of the Shapour river decreased by about 72%. Below normal annual precipitations of 2006 to 2018 and over-utilization of river catchment water resources during the same period are the main reasons for sharp decrease in stream ow of the Shapour river in recent years.
In order to illustrate the over-utilization of the upper Shapour river freshwaters resources, image of extensive water removal from the upstream Shapour river (Sasan spring) by water trucks is shown in Fig. 4.
To illustrate the in uence of recent prolonged stream ow reduction period on Shapour river salinity, mean annual discharge and salinity variation in hydrometric stations of the Shapour river catchment (stations S1, S2, S4, and S6) are shown in Fig. 5.
It follows from Fig. 5 that depending on the annual mean discharge value, salinity of the Shekastian salty tributary (station S2) has been uctuating around the average value of 10.03 dS m − 1 . Additionally, longterm variability is not observed in salinity data of this tributary. Before initiation of reduced stream ow period in 2006, Shapour river salinity levels at stations S1, S4 and S6 had been uctuating around average values of ~ 0.80, ~ 3.18, and 3.76 dS m − 1 , respectively. Compared to the period prior to stream ow reduction, average value of river salinity at station S1 increased by 20% and reached the value of ~ 0.96 dS m − 1 . River salinity at station S4 substantially increased over the stream ow reduction period, and during the 2017-2018 water year reached the maximum value of ~ 7.08 dS m − 1 (122% increase compared to prior to stream ow reduction period). Over the three initial years of stream ow reduction, river salinity at station S6 (downstream of dam) showed sharp increasing trend and during the 2008-2009 water year reached the maximum value of 6.90 dS m − 1 (82% increase compared to prior to stream ow reduction period). At the 4th year of low stream ow period, river salinity at station S6 sharply dropped and was then stabilized at the average value of 4.69 dS m − 1 , which is only 25% larger than that of pre-stream ow reduction period.
As Fig. 5 shows, salinity variation among the hydrometric stations of the Shapour river catchment are quite different. The reasons for differences are discussed next. The Shekastian salty tributary catchment is mountainous and its water resources have not been utilized. Therefore, due to lack of human intervention, tributary salinity was not nearly in uenced over time. In contrast, agricultural and municipal sectors in the upper Shapour river catchment were in a developing stage, causing increase in drainage water. While freshwater resources decreased due to low precipitation and over-utilization, the drainage water of developing agricultural and municipal sectors of the upper Shapour increased and consequently, the upper Shapour river salinity was intensi ed by ~ 20%. For stations S4 and S6, trend of salinity uctuation is strongly dependent on the operational schemes of the Shapour river water resources. As previously mentioned, most of freshwater resources of the Shapour river catchment are located above station S1. Due to water scarcity over the reduced stream ow period, water resources were over-utilized by means of water wells and direct water removal from river and springs. Consequently, freshwater ow of the Shapour river at station S1 was sharply decreased. In contrast, due to salinity, Boushigan brine spring and Shekastian salty tributary were not used. To show unbalanced utilization of water resources of Shapour river catchment, annual variation in the ratio of salty/brine out ow from Shakastian tributary and Boushigan spring to fresh-out ow from the Shapour river at station S1 are shown in Fig. 6. It is noted that due to constant discharge of ~ 11 l s − 1 , annual out ow volume from the Boushigan brine spring was calculated as the constant value of ~ 0.35 MCM. As shown in Fig. 6, ratio of Shakastian salty tributary out ow to the Shapour river fresh-out ow increased from 16.34% over 1990-2009 to 44.48% over 2009-2019 (approximately 3 times rising). Additionally, ratio of Boushigan brine spring out ow to the Shapour river fresh-out ow increased from 0.13% over 1990-2007 to 0.45% over 2007-2019 (more than 3 times rising). Therefore, ~ 3 times increase in the ratios of salty/brine out ow of Shekastian tributary and Boushigan spring to fresh-out ow of the Shapour river was the main reason for sharp increase in river salinity at station S4. Furthermore, ~ 20% rising in the Shapour river salinity at station S1 was in uential on river salinity rising at station S4.
Annual uctuation in Shapour river salinity at downstream of dam can be justi ed by calculation of longterm variability in the quota of relatively freshwater resources between stations S4 and S6 from the Shapour river ow. The mountainous nature of the catchment area between stations S4 and S6 has made it unsuitable for agricultural practices, and so it has mostly remained out of human reach. Except for Malicheh brackish tributary, other contributing water resources below station S4 act as the salinity dilutor for salty out ow water from station S4. Figure 7 shows annual variations in the ratio of water resources entering the Shapour river reach between stations S4 and S6 to the river out ow at station S6. Prior to dam operation, increased water into the Shapour river at the reach between stations S4 and S6 was calculated as the difference between the annual Shapour river out ows at stations S4 and S6. After that, it was calculated by subtracting the annual Shapour river out ow at station S4 from annual in ow water into the reservoir.
As it can be seen from Fig. 7, the average value of the ratio of increased water into the Shapour river from water resources between stations S4 and S6 to the Shapour river out ow at station S6 increased from 28.35% for the 1982-2009 period to 54.46% over the 2009-2019 period. While Shapour river out ow at station S4 was strongly decreased due to over-utilization of upstream water resources for irrigation and drinking purposes, water resources between stations S4 and S6 were not used. It resulted in ~ 2 times rising in the quota of fairly freshwater resources between stations S4 to S6 from the Shapour river ow and consequently, in Shapour river salinity reduction and stabilization below dam during 2009-2019. However, during the 2009-2019 period, some of river salinity rising at station S6 is due to surface evaporation from stored water in the reservoir. Based on measured daily pan evaporation data and daily reservoir water surface area, an average value of 12% of the reservoir in ow was lost through surface evaporation, contributing to an average increase of 12% in river salinity below the dam. Overall, Figs Based on presented data in Fig. 8, mean river salinity at the Khesht plain inlet showed sharp increase from ~ 2.85 dS m − 1 over 1982-1985 to ~ 6.7 dS m − 1 over 2016-2019 (139% rise). At the Khesht plain outlet, mean salinity increased from ~ 3.35 to ~ 6.5 dS m − 1 (97% rise). Smaller river salinity increase at the Khesht plain outlet indicates salt accumulation in the soil pro le and groundwater of the Khesht plain, due to using salty water of Shapour river for irrigation.
To investigate the in uence of the Raesiali-Delvari storage dam on the Shapour river salinity, measured river discharge and salinity values at the reservoir inlet and outlet (stations S5 and S6) are compared in Fig. 9. It is noted that during the period of data availability at the reservoir inlet (station S5), Shapour river salinity at station S4 was uctuating between 4.8 to 8.6 dS m − 1 .
It follows from Fig. 9 that the strong Shapour river salinity uctuation of 0.9-10.7 dS m − 1 at reservoir inlet (station S5) was adjusted to 3.6-5.5 dS m − 1 at reservoir outlet (station S6), which is attributed to annual mixing of stored water in the reservoir. Compared to station S4, the higher salinity uctuation of the Shapour river at station S5 is due to discharge from the Malicheh brackish tributary, located between stations S4 and S5. Despite 12% salinity rise due to surface evaporation from dam reservoir, mean river salinity below the dam (4.5 dS m − 1 ) is ~ 13.5% smaller than that of observed above the dam (5.2 dS m − 1 ). Dilution of stored water in the reservoir is the direct result of freshwater discharge by Tanbakoukar tributary and also direct precipitation over the lake area in the reservoir.
Due to density ow, the more saline in ows of reservoirs usually accumulate on the reservoir bottom levels. Considering abundance of saline water in ows into the Raeisali-Delvari reservoir (Fig. 9), stability of reservoir salinity and thermal strati cation and consequently, reservoir eutrophication would be possible. To check the possibility of reservoir stability in the Raeisali-Delvari dam, vertical variability of temperature and salinity of stored water in the reservoir were measured at two different times, i.e., 15-Jul-2017 and 21-Jan-2018, as shown in Fig. 10. The date of the 2nd measurement coincided with the occurrence of maximum salinity in the Shapour river upstream of the dam.
As Fig. 10 shows was not subjected to stability of reservoir strati cation. Annual cycle of reservoir strati cation and mixing con rms the fact that annual mixing of stored water in the Raeisali-Delvari reservoir is the main reason for salinity adjustment of the Shapour river in the reservoir downstream location.
The semi-arid climatic conditions of the Shapour river catchment have made it suitable for date farming. Based on classi cations presented by Maas and Hoffman (1977) and Maas (1993), date is a salt tolerant plant with threshold salinity (i.e., salinity at initial yield decline) of 4 dS m − 1 and 3.6% yield reduction per unit increase in salinity beyond threshold. Over the years, Shapour river salinity downstream of station S1 (below Shekastian salty tributary con uence) has always been a major problem and consequently, date farming has been the dominant agricultural crop in the middle and the lower Shapour river. As previously shown in Fig. 8, over the recent stream ow reduction period, Shapour river salinity at the Khesht plain inlet reached the maximum value of 26.0 dS m − 1 . Consequently, production of irrigated date orchards in the Khesht plain has decreased in terms of quantity and quality. Several date orchards have also become abandoned because of increase in salinity of irrigation water. For example, an image of the abandoned date orchards in the Khesht plain is shown in Fig. 11. In addition to salinity damages to the date orchards of the Khesht plain, the Shapour river water diversion and distribution facilities in the Khesht plain have also been damaged by being subjected to corrosion and destruction by river salinity (Fig. 12).
Based on the results of present study, Shapour river stream ow quality from con uence of Shekastian salty tributary to the Raeisali-Delvari reservoir inlet is in the worst condition in terms of the mean value and uctuation domain of salinity. Field observations over the recent stream ow reduction period indicated that approximately all of the aquatic species of the Shapour river at the above-mentioned reach have disappeared due to intense river salinity. Additionally, river water has not been suitable for riparian species, domestic animals, and irrigation of most of crop species. Therefore, the most severe economic costs and environmental damages of salinity rising occurred in the middle portion of Shapour river. Despite up to 140% mean salinity rising and strong river salinity uctuation in the middle Shapour, river salinity below the dam (lower Shapour river) is less intensi ed by recent stream ow reduction period. So far, salinity at the worst year of river salinity (2017-2018 water year) has been exceeded by as much as twice units of the threshold salinity value for date tree. Consequently, production quantity and quality of date orchards on downstream of the dam have minimally been in uenced by river salinity rising. Minimal damage caused by river salinity rising below the dam can be attributed to river salinity adjustment by annual mixing in the Raeisali-Delvari reservoir and reservoir water dilution by unused fairly freshwater resources between the Khesht plain outlet and the dam site. Success of the Raeisali-Delvari storage dam for salinity management of the Shapour river is due to its suitable location on the Shapour river, by being situated downstream of all of the main river tributaries with natural saline/fresh sources of water. Therefore, the reservoir can collect and mix all entering waters and by acting as a regulator, can produce large volume of water of reasonable quality.

Conclusions
River salinity rising over prolonged stream ow reduction periods is a major challenge for freshwater rivers by being subjected to natural salinity sources. Over the stream ow reduction periods, salinity of rivers with natural salinity sources are intensi ed by several parameters, including unbalanced use of river fresh/saline water resources, drainage water of irrigated agricultural elds with river water, and damming. Based on the ndings of present research, over-utilization of river freshwater resources during a stream ow reduction period was the main cause of dramatic rise in river salinity, due to increased ratio of saline waters to freshwaters. On the seasonal time-scale, drainage water of irrigated agricultural elds along with river ow could adjust river salinity uctuation by consecutive processes of salt accumulation during the irrigation season and then salt drainage during rainy seasons. Annual mixing of stored water inside the reservoir can result in adjustment of strong river salinity uctuation domain. Therefore, construction of storage dam on downstream of subjected rivers to natural salinity sources along with minimal utilization of fresh water resources in the upstream catchment can sharply decrease river mean salinity and its uctuation domain and consequently, providing more water of reasonable quality.

Declarations
This work was supported by the Iran Water Resources Management Company (IR), Regional Water Company of Fars [Grant number: FAW-97002]. The authors declare that they have no known competing nancial interests or personal relationships that could have appeared to in uence the work reported in this paper.   Image of extensive water removal from the upstream Shapour river (Sasan spring) by water trucks (image taken on 5-August-2018)  Annual variation in quota of Shekastian salty tributary and Boushigan brine spring from the Shapour river ow a) Ratio of salty-out ow from Shakastian tributary to fresh-out ow from the Shapour river at station S1 b) Ratio of brine-out ow from Boushigan spring to fresh-out ow from the Shapour river at station S1 Annual variation in the ratio of the increased water into the Shapour river from water resources between stations S4 and S6 to the Shapour river out ow at station S6  The Shapour river discharge and salinity uctuation at the Khesht plain inlet and outlet (stations S3 and S4) for 1982-1985 (prior to stream ow reduction period) and 2016-2019 (over stream ow reduction period) Figure 9 Salinity and discharge uctuation of the Shapour river at inlet and outlet of the Raeisali-Delvari reservoir  An image of the abandoned date orchards in the Khesht plain due to salinity rising of the Shapour river (image was taken at 5-August-2018) Figure 12 Images of corrosion and destruction of the Shapour river diversion and distribution facilities in the Khesht plain due to salinity rising of the Shapour river (image was taken at 5-August-2018)