Free Access

What are the Sustainable Water Policies in Central Regions of Iran? An Integrated Water Resource Management Model

Hossein Dastkhan

https://orcid.org/0000-0002-4042-2908

Department of Financial Engineering, Faculty of Financial Science, Kharazmi University, Tehran, Iran

E-mail Address: hdastkhan@khu.ac.ir

Corresponding author.

Search for more papers by this author

and

Golnaz Mohamadi

Department of Industrial Engineering, Imam Javad University College, Yazd, Iran

Search for more papers by this author

https://doi.org/10.1142/S2382624X23400015Cited by:3 (Source: Crossref)

This article is part of the issue:

Special Issue on Advances in Hydro-Economic Modeling for Sustainable Basin Management
Guest Editors: Alfonso Expósito, Taher Kahil and Julio Berbel

Abstract

The efficient and optimal management of water resources is of great importance due to the strong dependence of human life on water. Also, the availability and utilization of the existing water resources in regions with water shortages, such as the Middle East, impose high social, economic, and environmental costs. Therefore, water resource management policies should incorporate all aspects of supply, allocation, control, and monitoring of resources. This study provides a dynamic simulation model for water resource management in the Yazd province of Iran to examine the effect of different policies on the other variables and choose the most suitable policies over time. For this purpose, we used a system dynamics approach to propose an integrated water resources management model, considering a comprehensive view of different aspects. The proposed model included over 230 influential variables in water resources, along with economic, demographic, technological, agricultural, industrial, public policy-making, water demand-supply, and virtual water volume subsystems. After we validate the model, we define three scenarios (optimistic, baseline, and pessimistic) and four policy packages (i.e., business as usual, focusing on economic development, focusing on sustainable development based on ecological balance, and focusing on water conservation) in the time horizons of 10, 20, and 50 years. The simulation results indicated that we require a reform in the Yazd economic development strategies through policies such as changing the cropping pattern and reducing water-consuming industries. Moreover, water supply enhancement by raising the inter-basin transfer of water alone cannot be an effective solution for reducing the water shortage in Yazd province and may even worsen the water shortage in the long run. We conclude that the “Focus on Water Conservation” policy is the best solution to reduce water shortage. The results of the baseline scenario show that adopting the FWC policy changes the increasing trend of water shortage in the province and decreases them from more than 53 billion m³ in the “Business as Usual” policy to less than 20 billion m³ in the year 2040. As the FWC would decrease more than 10 billion m³ of the virtual water level imbalance compared to BAU policy and make more investments in water efficiency plans, it preserves the current resources of water in the long run.

Keywords:

1. Introduction

Water is a vital resource in human lives and activities, and it has long been a challenge to receive a reliable source of high-quality water. The ever-increasing water demand caused by population growth, industrialization, and living standard improvement has recently led to new and more complex problems, such as water supply-demand imbalance, drinking water issues, soil and water contamination, and environmental problems. It is required to predict the water resource systems and their users to tackle these growing problems. However, complexities in such systems and their uncertainties impose some difficulties in understanding their behaviors. Water resource experts are searching for new models to make more effective decisions for sustainable development. These models should require incorporating the complex relationships among different factors, their uncertainties in nature, and the interactions of their related subsystems. In general, we applied water resource management models to make optimal and sustainable use of water resources to make the maximum response to demands. In other words, the relationships between supply and demand and economic advantages should be incorporated into an integrated modeling framework for the efficient allocation of water resources (Mirchi et al., 2012).

The Middle East has always encountered water tension due to low precipitation. As a developing country, Iran has experienced changes in the water pattern for 50 years and has focused on making maximum and unplanned use of water resources, as with many other countries. Lying on the arid desert belt of the Northern Hemisphere, Yazd Province is among the driest provinces in Iran. The warm and dry climates of Yazd Province arise from its high distance from water bodies (e.g., seas), high solar radiation and evaporation, and topographic conditions. Furthermore, the improper use of new water extraction technologies and unsuitable water governance has led to the quality and quantity deterioration of water resources in Yazd Province in the past 50 years. Along with climate change, these factors have resulted in concerns regarding water resources (Statement No. 2 of the Yazd water perspective, 2020).

Water resource management at a regional scale under climate change is complex due to different aspects. At first, these systems are large-scale with large spatial extensiveness due to the number of components and the complex relationships between them. Moreover, most of these relationships are expressed by nonlinear and, sometimes, discontinuous and non-convex functions. The time-variance and dynamism of most variables, such as water demand, water supply, virtual water level, groundwater level changes, and the population is another aspect of water resource management complexity. These systems are also complex due to the necessity to model different economic, social, political, and environmental sectors. Lack of details and inaccurate measurements of variables and parameters, such as the percentage of water returning to groundwater reservoirs (aquifer recharge) and the illegal well diggings, increase the system’s complexity. The water resource systems are multi-purpose at the regional or urban scale, where they try to achieve social satisfaction besides having sustainable water resource management while considering the economic and environmental protection objectives. Since these objectives are often inconsistent or even opposite, they make water resource management systems more complex. Another complexity of these systems relates to handling variables and relationships in the form of criteria that are measurable by standard instruments and techniques. Hence, due to the temporal and spatial complexity of economic, social, political, and environmental processes and water resource management, the methods that are employed to solve the problem would encounter complexities, considerations, and challenges. Due to nonlinearity and nonconvexity, such challenges make it difficult to develop suitable formulations and solve an efficient model that could suit the requirements of the problem (Mehr Azar et al.2016; Sterman 2000).

The literature review indicated that different approaches are available to model water resources, most of which have advantages and disadvantages. The optimization models have linear or nonlinear inflexible objective functions and constraints which do not capture many details. Statistical methods such as Monte Carlo simulation models are based on statistical data only and cannot represent interactions between components (Mirchi et al., 2012). Equilibrium models and Game theory can help find the general equilibrium of the system; however, they would not consider a large part of the details. Agent-based models have a bottom-up approach and are computational frameworks simulating the actions and interactions between agents. They may be suitable for water resource modeling. System dynamics is an agent-based model for modeling different systems and analyzing policies based on feedback system theory (Zomorodian et al.2018). In this paper, we employed system dynamics in this study for several reasons.

Lack of system thinking is probably the main explanation for the water tension in Yazd Province. Unplanned and instant decisions without considering their long-term consequences have resulted in many problems, including the water demand-supply imbalance. System dynamics and systematic thinking help realize the effects of policies over time and prevent measures that would impose adverse long-term impacts on water resources. Moreover, several factors are involved in water resource management, each with nonlinear effects that could reveal in the long run. System dynamics can define these variable relationships and complexities. These models can take quantitative values and simulate nonlinear objective functions and constraints. Since system dynamics can represent the actions, interactions, and effects of variables on the system, it is not limited to specific problems and can model nonlinear objective functions and constraints. Since several beneficiaries, such as the political, economic, and social sectors, are involved in the Yazd water resource system, it is necessary to adopt a novel approach to simulate and analyze the complexities between these sectors and represent the effects of each event over time. Thus, we choose system dynamics among other methods (Langarudi 2019).

This study defined water shortage as a problem to develop a system dynamics model of the water resources of the Yazd Province. The system dynamics modeling of water resources is an interdisciplinary problem requiring different expertise like economics, agriculture, environment, and civil engineering. Therefore, we acquired knowledge about the stakeholders and the influencing factors by reviewing relevant literature and interviewing experts. To model the water resources and socio-economic system of Yazd Province, we propose a dynamic hypothesis to explain why and how the problem appeared. Then, we developed a causal loop and the stock-flow diagram based on VENSIM and introduced the mathematical equations representing the relationships between the variables. In the last phase of the modeling process, the model was validated from 2011 to 2018 with different system dynamics validation methods.

The results show the suitable performance of the model in the water system simulation. We considered three different time horizons of 10, 20, and 50 years to find the capability of the proposed model in policy-making and predicting system behavior. We executed the simulation model for the time horizons using three scenarios and four policy packages and analyzed the sensitivity of each variable to change in other variables. The results could help the water policymakers in Yazd Province to make optimal and sustainable policies.

We can examine the contributions of this paper from two aspects. Although some previous attempts used system dynamics (SD) to analyze water resource management in the Middle East and Iran, they did not include a comprehensive study of all related subsystems. In other words, the previous similar studies in this region did not consider the population, water resources supply, investment in water-intensive and low-water-intensive industries, adjustment of agricultural methods, and other economic, social, environmental, and policy-making subsystems altogether. As a result, this paper is a unique study to analyze different policies to improve water resource management in the Middle East, especially in the central regions of the Iranian Plateau. On the other hand, and in comparison with other studies that have used SD for water resources modeling in other areas, as far as our knowledge, this paper is the first study that provides a comprehensive model which deals with the concept of virtual water level and the inter-basin water transfer along with other existing subsystems.

The remainder of the study is organized as follows: Section 2 reviews the system dynamics literature; Section 3 describes the system dynamics steps and model construction; Section 4 analyzes the model under the predefined scenarios and policies in different time horizons; and, Section 5 concludes the work.

2. Literature Review

The flexibility and accuracy of the System Dynamics method help identify the effects of factors on a system when the system is affected by socio-economic factors (Xu et al.2002). Moreover, several studies have focused on adopting SD in the analysis of different socio-economic systems in various fields in the past two decades, including energy, environment, water, and sociology (e.g., see Bagheri and Hosseini 2011; Bawary et al.2022; Dastkhan and Owlia 2014; Madani 2007; Owlia and Dastkhan 2012; Yang et al.2008 2014). In particular, a large amount of literature has employed System Dynamics to analyze water resource management systems.

In a leading paper, Stave (2003) tried to increase public knowledge of water resource protection in Las Vegas. She adopted SD to demonstrate the outcomes of different strategies in this area. She selected a total of 83 interviewees ages 18–65 years, including teachers, students, environmental experts, and retired citizens, and asked them to propose measures to avoid the water shortage in Las Vegas. They suggested policies that included water resources enhancement, reducing the water consumption of hotels, reducing the domestic water consumption of citizens, reducing the external water consumption of citizens, reducing immigration to the state, and combining all the policies. The simulation results showed that decreasing external consumption had the best results. Ahmad and Simonovic (2004) developed the Canadian integrated water management model. They proposed 12 scenarios and evaluated their impacts on freshwater availability, economic growth, population growth, energy generation, and food production. The results revealed that the future development of Canada would be significantly dependent on the acceptable quality of water resources and the control of water consumption in different sectors.

Xi and Poh (2013) represented a System Dynamics simulation of sustainable water management in Singapore. Singapore has invested in water desalination, wastewater reclamation, basin management, and other similar projects to achieve a self-sufficient water system. They developed the SD model to analyze the long-term outcomes of different investment schemes. They concluded that investing in groundwater reservoirs or surface basins would not solely be sufficient to achieve self-sufficiency. Balali and Viaggi (2015) modeled groundwater resources through the system dynamics approach under different economic and climate change scenarios. They studied 16 irrigation and energy pricing policies under arid, moderate, and humid climates. They changed the prices of water, energy, and agricultural products and the irrigation levels for different policies. Comparing these policies with the current policy indicated that a rise in the water price would comparatively decrease groundwater extraction.

Madani (2007) adopted the system dynamics approach to realize whether aquifer water extraction was the best solution to the challenge of supplying water to different consumer groups in the Zayanderud Basin, Iran. The simulation results of various scenarios demonstrated how variables influenced the system. He found the most desirable outcome in controlling groundwater extraction was for those scenarios controlling population growth and water demand.

Bagheri and Hosseini (2011) studied the water resources, especially the groundwater resources of Mashhad Plain, Iran. They found that the region’s development was strongly dependent on agriculture, despite the reduced levels of surface water resources. They employed statistical data and demonstrated that the agriculture sector had the highest water consumption but the lowest contribution to the value-added GDP of the region. They attempted to identify the roots of the problem and propose sustainable strategies to mitigate them by the System Dynamics approach. They proposed two sustainability strategies (1) economic growth based on the water resource limitations and (2) water resource allocation based on the value added. For these strategies, several policies were proposed, including enhancing the wastewater reuse system, changing the sharing of agricultural water, and changing the cropping pattern. The simulation results in Vensim revealed that changing the cropping pattern of wheat would be the most effective policy.

Gohari et al. (2013) exploited System Dynamics modeling and investigated the policy of intra-basin water transfer to supply the increasing water demand in the Zayanderud Basin, Iran. They demonstrated that intra-basin water transfer would be an insufficient, unsustainable policy in water resource management and impose unwanted adverse impacts. The results showed that a suitable, solid solution included a policy package that controlled system behavior and minimized the likelihood of undesirable outcomes. More precisely, policymakers should consider the water demand reduction for agricultural consumption by changing the cropping pattern along with intra-basin water transfer. In another paper, Gohari et al. (2017) proposed a system dynamics model to analyze the adaptation strategies related to water resources and biophysical and socio-economic subsystems in the Zayandeh River area of Isfahan. They showed that infrastructure improvements, careful water demand management, and ecosystem-oriented regulations with supply enhancement plans could temporarily reduce water tension in this area. The proposed model of this study considered the agricultural, hydrological, and socio-economic subsystems.

Shahbazbeygan (2016) conducted a dynamic analysis of land-use planning in the water resources of Southeast Iran and the Hirmand River. Zomorodian et al. (2018) conducted a comprehensive review of the System Dynamics literature and suggested that System Dynamics could provide a deep insight into interactions and feedback between interconnected subsystems of a water resource system. Thus, it was a suitable instrument to handle the complex water resource relationships. They argued that System Dynamics has some limitations. It needs a detailed and comprehensive understanding of the system and has a limited capacity to handle the high uncertainty level of such systems. It is also a simulation technique and cannot always yield proper outcomes. They proposed that combination of System Dynamics with other methods, such as deep neural networks (DNNs), game theory, agent-based simulation, fuzzy logic, or the integration of System Dynamics with optimization techniques, such as the Krill-Herd algorithm (KA) or genetic algorithms (GAs) would be beneficial in identifying optimal policies and decision making. In another paper, Langarudi (2019) introduced a system dynamic model of water problems in the arid and semi-arid areas of New Mexico and proposed solutions.

Layani et al. (2021) applied the system dynamic modeling to analyze the Kowsar Dam Basin in southwest Iran, which is facing a critical challenge due to periodic drought. They stated that the change in water resource access has a dynamic behavior over time under the influence of many factors like population growth and climate change. They showed that water supply and demand imbalance could cause significant problems in this area. The government should educate people to control their daily water consumption to solve the problem of water shortage in a good way. Their proposed model has considered the variables related to population variables, water hydrology, and weather climate change.

This study adopts a systemic approach to analyze the whole aspects of the water crisis of Yazd Province, Iran. We identified the key factors and their mutual relationships over time and simulated the results SD model to provide a suitable policy-making instrument. The proposed comprehensive model incorporates the economic, social, and virtual water subsystems and enables the visualization of the effects of different scenarios and policies over time. Thus, the model would be helpful in large-scale public decision-making in the region and could prevent or minimize improper policy-making with destructive impacts on natural resources.

3. Case Study

3.1. Environmental conditions of Yazd province

Yazd Province lies in an arid and hot climate in the northern hemisphere. The dry and hot weather of Yazd arises from its high distance from water bodies (e.g., seas), intense solar radiation, high level of evaporation, and topographic conditions. Also, the improper use of new water extraction technologies and the lack of good water governance have led to quantitative and qualitative challenges in the province’s water resources for the past 50 years. The combination of climate changes with these factors has put the province’s water resources at a high level of risk (Statement No. 2 of the Yazd water perspective 2020). We will discuss these issues further.

We utilized the historical data from 2011 to 2018 to realize the current status and the future perspective of these problems. Figure 1 illustrates the model behavior diagrams of critical variables related to the Yazd water resources from 2011 to 2018. Figure 1 shows that the demand for industrial, mine, and service sectors increased during 2011–2018. We expect that the IMS demand will continue to increase due to the new investments in the IMS sectors of Yazd Province. Domestic water demand will also increase due to population growth and the high immigration of job-seeking workers in the province. As can be seen in Figure 1, the water demand of the agricultural sector declined during these years due to the reduced capacities of groundwater resources and land-use changes. We expect that the trend will continue in the future.

According to the industrial sector, the annual manufacturing rates of raw steel and related productions and ceramic tiles, as water-consuming industries, increased during 2011–2018. The evidence shows that the capacity development of water-consuming industries will continue to rise. In the agricultural sector, fruits, vegetables, and dairy production experienced significant growth. As a result, it can be inferred from Figure 1 that the water demand of Yazd Province is likely to rise in the future. Therefore, proper management of water resources is necessary due to supply shortages and the negative groundwater balance.

Since 1998, the Kowsar water transfer project has provided water from the Zayandehroud river basin with a capacity of 67Mm³ per year. Currently, the Persian Gulf water transfer project to Yazd Province is considered to supply the Yazd required water by the industry and mining sector with an annual capacity of 20Mm³. Nevertheless, due to the dry climate, the growing industrial development, the volume of new investments, and significant migration to this province, Yazd has always faced a significant deficit in the supply of water resources.

4. Methodology

4.1. Dynamic hypothesis

Considering the behavior of the water shortage in Yazd Province, we assume that a rise in the water transfer would increase the water supply and moderate the water shortage in Yazd Province. However, since an increase in the water supply will associate with economic growth and may result in worsened water shortage, it is required to investigate the water demand elevation and economic development simultaneously. This work has focused on this as a dynamic hypothesis. Figure 2 depicts the conceptual model based on the dynamic hypothesis.

Figure 2 shows the initial dynamic hypothesis of our SD model. The dynamic hypothesis states that with the increase in water transfer to the province, the water supply will increase and adjust the province’s water shortage. However, since the water supply increase can lead to economic development, it is necessary to examine the issue of increasing demand with increased economic activities. These effects may ultimately aggravate the water shortage in the province.

Once we developed the dynamic hypothesis and model boundary graph, we identified the main subsystems and their relationships and presented them as a subsystem diagram. These subsystems are agricultural water demand, industrial water demand, domestic water demand, demographic variations, water supply, water sector’s financial resources, water technology, virtual water level, public policy-making, and water economy.

4.2. Causal loop diagram

•	Figure 3 depicts the causal loop diagram with several positive and negative causal loops. It represents an overall analysis of the causal behavior of the system variables and their interactions in the form of positive (reinforcing feedback) and negative (balancing feedback) loops. However, we have elaborated on the main loops that have an important role in the water system behavior, as follows:
•	Water Supply Positive Loop (R1): A rise in the water supply increases the tendency of investors to invest in water-consuming industries in the long run, increasing industrial water consumption. It can increase water demand and consequently reduce water resources. Moreover, it also can increase the tendency to invest in wastewater treatment systems due to elevated wastewater. As a result, the surface water will increase resulting from wastewater treatment enhancement and enhances the province’s water supply to some extent.
•	Inter-Basin Transfer Positive Loop (R₂): The water transfer to the province increases when water demand exceeds the water supply. It induces an illusion of water abundance among the public and investors, raising the tendency to invest in water-consuming industries. For example, large and small steel companies around Yazd Province have experienced significant growth in the past decade.
•	Water Supply Negative Loop (B₁): Increased water resources in the province raise the tendency to invest in water-consuming industries. Therefore, it leads to new employment opportunities, increases the rate of immigration which leads to population growth, increases domestic water consumption, and consequently increases the total water demand. As a result, the effect of water resource enhancement moderates with delay.
•	Water Shortage and Agriculture Reforms Negative Loop (B₂): A reduction in water resources enhances governmental policies toward reducing agricultural water consumption. Such measures are costly and require investment for technology enhancement. The increase in the application of new technologies, such as drip and pressurized irrigation systems, will enhance the efficiency of the agricultural sector, diminishes agriculture water consumption, and increases water supply.
•	Water Shortage and Industrial Reforms Negative Loop (B₃): Water resource reduction directs industrial development policies toward the development of low-water-demand industries. The water consumption of the industrial sector declines as the tendency to invest in low-water-demand industries rises, thereby enhancing water resources.
•	Inter-Basin Transfer and Virtual Water Negative Loop (B₄): As water supply and the tendency to invest in water-consuming industries increase, production exceeds consumption, raising exports in the provinces. It diminishes the virtual water level and water supply. Tile and steel companies are among such industries.

4.3. Stock and flow diagram

Figure 4 shows the stock and flow diagram of the System Dynamics model. This diagram tries to analyze the relationships between components with mathematical equations. The SFD is directly related to the causal loop diagram presented in the previous section. However, they cannot completely match each other due to technical differences. Our proposed stock and flow diagram involve 12 stock variables. We described the main subsystems of the proposed SFD as follows:

— Water demand subsystems: Water demand in Yazd Province divides into domestic, industrial-mineral-service (IMS), and agricultural water demands. The gross domestic product (GDP) and water consumption intensity of each sector are the main factors influencing the water demands of these sectors. Domestic water demand refers to drinking water consumed in houses and religious places. We collect the domestic water demand data from the regional water statistics center and statistical yearbook of Yazd Province (Statistics Center of Iran 2010–2019).

As mentioned before, domestic water demand is a function of domestic sector GDP and the domestic water consumption intensity.

Domestic water demand = domestic water consumption Intensity * Domestic sector GDP . <math display="block" altimg="eq-00003.gif"><mtable columnalign="left" displaystyle="true"><mtr><mtd><mrow><mstyle><mtext mathvariant="normal">Domestic water demand</mtext></mstyle><mo>=</mo><mstyle><mtext mathvariant="normal">domestic water consumption Intensity</mtext></mstyle></mrow></mtd></mtr><mtr><mtd><mspace width="10pc"></mspace><mo>*</mo><mspace width=".17em"></mspace><mstyle><mtext mathvariant="normal">Domestic sector GDP</mtext></mstyle><mo>.</mo></mtd></mtr></mtable></math> (1)

The population is also a factor that impacts domestic water consumption. The population changes are a function of the annual growth and immigration rates. Immigration to Yazd Province varies based on employment opportunities. The other main equations are provided in Appendix A.

Population (t) = Population (0) + \int t 1 (Population growth + Immigration) * d t, <math display="block" altimg="eq-00004.gif"><mstyle><mtext mathvariant="normal">Population</mtext></mstyle><mo stretchy="false">(</mo><mi>t</mi><mo stretchy="false">)</mo><mo>=</mo><mstyle><mtext mathvariant="normal">Population</mtext></mstyle><mo stretchy="false">(</mo><mn>0</mn><mo stretchy="false">)</mo><mo>+</mo><msubsup><mrow><mo>\int</mo></mrow><mrow><mn>1</mn></mrow><mrow><mi>t</mi></mrow></msubsup><mo stretchy="false">(</mo><mstyle><mtext mathvariant="normal">Population growth</mtext></mstyle><mo>+</mo><mstyle><mtext mathvariant="normal">Immigration</mtext></mstyle><mo stretchy="false">)</mo><mo>*</mo><mi>d</mi><mi>t</mi><mo>,</mo></math> (2)

Immigration = Population * Immigration Rate . <math display="block" altimg="eq-00005.gif"><mstyle><mtext mathvariant="normal">Immigration</mtext></mstyle><mo>=</mo><mstyle><mtext mathvariant="normal">Population</mtext></mstyle><mo>*</mo><mstyle><mtext mathvariant="normal">Immigration Rate</mtext></mstyle><mo>.</mo></math> (3)

— Agricultural water demand subsystems: Agricultural water demand refers to the amount of water extracted from wells, springs, and subterranean canals in the agriculture sector. GDP and water consumption affect agricultural water demand as follows:

Net change of Agricultural GDP = Agricultural GDP * Agricultural GDP growth rate . <math display="block" altimg="eq-00006.gif"><mtable columnalign="left" displaystyle="true"><mtr><mtd><mrow><mstyle><mtext mathvariant="normal">Net change of Agricultural GDP</mtext></mstyle></mrow></mtd></mtr><mtr><mtd><mspace width="2pc"></mspace><mo>=</mo><mstyle><mtext mathvariant="normal">Agricultural GDP</mtext></mstyle><mo>*</mo><mstyle><mtext mathvariant="normal">Agricultural GDP growth rate</mtext></mstyle><mo>.</mo></mtd></mtr></mtable></math> (4)

Equation (5) shows that available financial resources in the water sector are the resources allocated by the government to the water sector, the revenues from selling water, and other financial resources. We considered consuming these resources at specific rates in different water development plans as Equation (6) (Shannak et al.2018). We obtained the agricultural water demand data from the regional water statistics center and the statistical yearbook of Yazd Province (Statistics Center of Iran 2010–2019). Figure 5 represents the stock and flow diagram of the agricultural sector.

Available Funds = Governmental Budget + Annual Revenues + Previous Residual Funds, <math display="block" altimg="eq-00007.gif"><mtable columnalign="left" displaystyle="true"><mtr><mtd><mrow><mstyle><mtext mathvariant="normal">Available Funds</mtext></mstyle><mo>=</mo><mstyle><mtext mathvariant="normal">Governmental Budget</mtext></mstyle><mo>+</mo><mstyle><mtext mathvariant="normal">Annual Revenues</mtext></mstyle></mrow></mtd></mtr><mtr><mtd><mspace width="7pc"></mspace><mo>+</mo><mspace width=".17em"></mspace><mstyle><mtext mathvariant="normal">Previous Residual Funds</mtext></mstyle><mo>,</mo></mtd></mtr></mtable></math> (5)

Funds for developing projects = Available Funds * Developing Funds consumption Rate . <math display="block" altimg="eq-00008.gif"><mtable columnalign="left" displaystyle="true"><mtr><mtd><mrow><mstyle><mtext mathvariant="normal">Funds for developing projects</mtext></mstyle></mrow></mtd></mtr><mtr><mtd><mspace width="2pc"></mspace><mo>=</mo><mstyle><mtext mathvariant="normal">Available Funds</mtext></mstyle><mo>*</mo><mstyle><mtext mathvariant="normal">Developing Funds consumption Rate</mtext></mstyle><mo>.</mo></mtd></mtr></mtable></math> (6)

Figure 5. Stock and Flow Diagram of Agricultural Water Demand

— IMS water demand: IMS water demand is the demand for drinking water, water transfer, or groundwater in the industrial, mineral, and service sectors. It is the total water consumption of industries, mines, public organizations, businesses, service companies, and the construction sector in Yazd Province. The GDP of the IMS sector and the water consumption intensity influences the IMS water demand as follows :

IMS Water Demand = IMS GDP \times IMS Water Consumption Intensity, <math display="block" altimg="eq-00009.gif"><mstyle><mtext mathvariant="normal">IMS Water Demand</mtext></mstyle><mo>=</mo><mstyle><mtext mathvariant="normal">IMS GDP</mtext></mstyle><mo>\times</mo><mstyle><mtext mathvariant="normal">IMS Water Consumption Intensity</mtext></mstyle><mo>,</mo></math> (7)

IMS GDP (t) = IMS GDP (0) + \int t 1 Net Change of IMS GDP * d t . <math display="block" altimg="eq-00010.gif"><mstyle><mtext mathvariant="normal">IMS GDP</mtext></mstyle><mo stretchy="false">(</mo><mi>t</mi><mo stretchy="false">)</mo><mo>=</mo><mstyle><mtext mathvariant="normal">IMS GDP</mtext></mstyle><mo stretchy="false">(</mo><mn>0</mn><mo stretchy="false">)</mo><mo>+</mo><msubsup><mrow><mo>\int</mo></mrow><mrow><mn>1</mn></mrow><mrow><mi>t</mi></mrow></msubsup><mstyle><mtext mathvariant="normal">Net Change of IMS GDP</mtext></mstyle><mo>*</mo><mi>d</mi><mi>t</mi><mo>.</mo></math> (8)

Additionally, investment in low- and high-water-demand industries have significantly influenced IMS water demand. These investments in the last years haven’t considered their consequences. The government permitted the private sector to establish water-consuming industries (e.g., the steel and tile industries) to achieve economic development. However, the disadvantages and consequences of such investments reveal now. Unfortunately, the moderation measures would worsen these consequences. It is a positive and reinforcing loop arising from the capacity enhancement of these industries.

Figure 6 illustrates the stock and flow diagram of IMS water demand. As can be seen, the transfer of water to Yazd Province affects investment in water-consuming industries. Based on the previous experiences of the Kowsar Inter-Basin water transfer project transferring water from the Zayandehroud basin to Yazd, we expect that increased water transfer to Yazd Province would increase investments in industries and thus increases the IMS water demand (Statement No. 2 of the Yazd water perspective 2020). Moreover, the increased employment opportunities increase immigration to Yazd Province, which impacts domestic water demand (see Eqs. (9) and (10)).

Change in Inter-basin Water Transfer = Water inlet from Kowsar Basin + Water inlet from other Basin + Water outlet from Yazd Basin, <math display="block" altimg="eq-00011.gif"><mtable columnalign="left" displaystyle="true"><mtr><mtd><mrow><mstyle><mtext mathvariant="normal">Change in Inter-basin Water Transfer</mtext></mstyle></mrow></mtd></mtr><mtr><mtd><mspace width="1.5pc"></mspace><mrow><mo>=</mo><mstyle><mtext mathvariant="normal">Water inlet from Kowsar Basin</mtext></mstyle><mo>+</mo><mstyle><mtext mathvariant="normal">Water inlet from other Basin</mtext></mstyle></mrow></mtd></mtr><mtr><mtd><mspace width="2pc"></mspace><mspace width="1em"></mspace><mo>+</mo><mspace width=".17em"></mspace><mstyle><mtext mathvariant="normal">Water outlet from Yazd Basin</mtext></mstyle><mo>,</mo></mtd></mtr></mtable></math> (9)

Inter-basin Water Transfer (t) = Inter-basin Water Transfer (0) + \int t 1 Change in Inter-basin Water Transfer * d t . <math display="block" altimg="eq-00012.gif"><mtable columnalign="left" displaystyle="true"><mtr><mtd><mrow><mstyle><mtext mathvariant="normal">Inter-basin Water Transfer</mtext></mstyle><mo stretchy="false">(</mo><mi>t</mi><mo stretchy="false">)</mo></mrow><mrow><mo>=</mo><mstyle><mtext mathvariant="normal">Inter-basin Water Transfer</mtext></mstyle><mo stretchy="false">(</mo><mn>0</mn><mo stretchy="false">)</mo></mrow></mtd></mtr><mtr><mtd><mspace width="2pc"></mspace><mspace width="1em"></mspace><mo>+</mo><mspace width=".17em"></mspace><msubsup><mrow><mo>\int</mo></mrow><mrow><mn>1</mn></mrow><mrow><mi>t</mi></mrow></msubsup><mstyle><mtext mathvariant="normal">Change in Inter-basin Water Transfer</mtext></mstyle><mo>*</mo><mi>d</mi><mi>t</mi><mo>.</mo></mtd></mtr></mtable></math> (10)

— Water supply of Yazd Province: The water supply of Yazd Province refers to the total water input of the Yazd. We classify the water supply into groundwater, water transfer, surface water and runoff, and virtual water (see Eq. (11)). We consider the groundwater supplies as the water resources aggregated in saturated layers under the ground. Water transfer represents the total amount of water transferred to the province through inter-basin transfer. Surface water and runoff include permanent rivers and post-rainfall streams. Virtual waters are the amount of water consumed by industrial, agricultural, livestock, and poultry products in the production process (Mokhtar and Aram 2017). It is the total amount of water embodied in different stages of the production cycle. Figure 7 indicates the stock and flow diagram of the water supply in Yazd Province.

Water Supply = final groundwater amount + inter-basin water transfer + normal water extraction + IF THEN ELSE (net virtual water level > 0, net virtual water level, 0) . <math display="block" altimg="eq-00013.gif"><mtable columnalign="left" displaystyle="true"><mtr><mtd><mrow><mstyle><mtext mathvariant="normal">Water Supply</mtext></mstyle><mo>=</mo><mstyle><mtext mathvariant="normal">final groundwater amount</mtext></mstyle><mo>+</mo><mstyle><mtext mathvariant="normal">inter-basin water transfer</mtext></mstyle></mrow></mtd></mtr><mtr><mtd><mspace width="6pc"></mspace><mrow><mo>+</mo><mspace width=".17em"></mspace><mstyle><mtext mathvariant="normal">normal water extraction</mtext></mstyle></mrow></mtd></mtr><mtr><mtd><mspace width="6pc"></mspace><mrow><mo>+</mo><mspace width=".17em"></mspace><mstyle><mtext mathvariant="normal">IF THEN ELSE</mtext></mstyle><mo stretchy="false"> (</mo><mstyle><mtext mathvariant="normal">net virtual water level</mtext></mstyle><mo>&gt;</mo><mn>0</mn><mo>,</mo></mrow></mtd></mtr><mtr><mtd><mspace width="6.5pc"></mspace><mstyle><mtext mathvariant="normal"> net virtual water level</mtext></mstyle><mo>,</mo><mn>0</mn><mo stretchy="false">)</mo><mo>.</mo></mtd></mtr></mtable></math> (11)

We should note that the water supply of Yazd Province increases when the virtual water input is greater than the virtual water output. Otherwise, a negative virtual water level would be a demand factor and reduce the water supply. The water shortage in the province is the difference between water demand and water supply. The remaining equations are provided in Appendix A.

Water Shortage = Water Supply - Water Demand . <math display="block" altimg="eq-00014.gif"><mstyle><mtext mathvariant="normal">Water Shortage</mtext></mstyle><mo>=</mo><mstyle><mtext mathvariant="normal">Water Supply</mtext></mstyle><mo>-</mo><mstyle><mtext mathvariant="normal">Water Demand</mtext></mstyle><mo>.</mo></math> (12)

Figure 6. Stock and Flow Diagram of IMS Water Demand

Figure 7. Stock and Flow Diagram of Water Supply in Yazd Province

— Cost of Water and Water Tarrifs: The cost of water is the total cost from water extraction to water delivery to consumers. It involves digging, transferring, chlorination, treatment, and labor costs. The water tariff is the price paid by consumers for water. Figure 8 plots the stock and flow diagram of the cost and profit of water sales (see Eqs. (13) and (14)).

IMS Water Total Cost = Marginal Cost for IMS Water * IMS Water Demand, <math display="block" altimg="eq-00015.gif"><mstyle><mtext mathvariant="normal">IMS Water Total Cost</mtext></mstyle><mo>=</mo><mstyle><mtext mathvariant="normal">Marginal Cost for IMS Water</mtext></mstyle><mo>*</mo><mstyle><mtext mathvariant="normal">IMS Water Demand</mtext></mstyle><mo>,</mo></math> (13)

IMS Water Profit = (IMS Water Demand * Marginal Cost for IMS Water) * (1 + Profit margin of IMS Water) . <math display="block" altimg="eq-00016.gif"><mtable columnalign="left" displaystyle="true"><mtr><mtd><mrow><mstyle><mtext mathvariant="normal">IMS Water Profit</mtext></mstyle><mo>=</mo><mo stretchy="false">(</mo><mstyle><mtext mathvariant="normal">IMS Water Demand</mtext></mstyle><mo>*</mo><mstyle><mtext mathvariant="normal">Marginal Cost for IMS Water</mtext></mstyle><mo stretchy="false">)</mo></mrow></mtd></mtr><mtr><mtd><mspace width="7.5pc"></mspace><mo>*</mo><mspace width=".17em"></mspace><mo stretchy="false">(</mo><mn>1</mn><mo>+</mo><mstyle><mtext mathvariant="normal">Profit margin of IMS Water</mtext></mstyle><mo stretchy="false">)</mo><mo>.</mo></mtd></mtr></mtable></math> (14)

Figure 8. Water Cost Price and Sales Price Stock and Flow Diagram

4.4. Scenarios and policies

The definition of efficient scenarios requires the identification of exogenous variables of the water resource system in the province. This study incorporated 10 important exogenous variables based on which we extracted the number of scenarios. As it is clear from Table 1, we consider three different scenarios: baseline scenario, optimistic scenario, and pessimistic scenario. In the baseline scenario, we assumed that the existing conditions of our external variables would continue on the horizon. For the optimistic and pessimistic scenarios, we applied the worst and the best available historical values for each variable in the past 25 years.

**Table 1. Scenarios Proposed in the Yazd Water Resources System (Ministry of Jihad & Agriculture 2010–2020; Yazd Management and Planning Organization 2015–2020)**
	Scenario 1	Scenario 2	Scenario 3
	Optimistic Scenario	The Baseline Scenario	Pessimistic Scenario
Inflation	10%	15%	25%
Population growth rate	1%	1.30%	1.80%
Immigration rate to Yazd Province	0.30%	0.60%	0.80%
Budget growth rate	40%	22%	$-$ 40%
Inter-basin water transfer to Yazd Province (percentage change in Kowsar project)	5%	0%	$-$ 5%
Precipitation (mm)	130	100	70
Investment growth rate in industry, mining and services	40%	25%	0%
GDP growth rate of industry, mining and services	0.45%	0.28%	0.05%
GDP growth rate of the agricultural sector	0.45%	0.35%	0.00%
GDP growth rate of the domestic sector	0.40%	0.26%	0.05%

Along with exogenous system variables that are not under the control of water decision-making authorities, there are several endogenous variables in the system. Analyzing the effects of such endogenous variables helps to obtain deeper insights into the system and its outcomes. As a result, along with the definition of scenarios, this study analyzed water resource policies for Yazd Province. The endogenous variables were developed as policy packages, discussing their effects on the water resource system while measuring the impacts of different scenarios.

Table 2 reports the proposed policy packages. Table 3 provides the quantities of the variables in the defined policies. The considered values for different policy packages are determined based on available historical data and expert opinions. For each scenario, we analyzed the set of policy packages to identify the optimal policy with the best outcome.

**Table 2. Policies Proposed in the Yazd Water Resource System**
Policy	Description
Business as Usual (BAU)	This policy suggests preserving the current conditions
Focus on Economic Development (FED)	This policy proposed the maximization of the economic indicators in the province. The government will grant more subsidies to bring water prices to a satisfactory level for all users so that economic development can continue by expanding different investments. The first version of this policy (FED1) assumes that increased water demand arising from economic development will meet by increasing the extraction of groundwater resources. However, the second version of this policy (FED2) increases water transfer to the province by 250 million cubic meters per year, in addition to the Kowsar Basin. A significant portion of the transferred water will serve the industrial sector. Moreover, one-fifth (50Mm³) will dedicate to the agriculture sector. Thus, increased water supply in this policy will lead to economic development in the IMS and agricultural sectors of the province
Focus on Sustainable Development (FSD) based on Environmental Balance Conservation	In this policy, the policy-makers primarily focus on sustainable development based on environmental protection. Therefore, this policy eliminates domestic water subsidies and reduces industrial water subsidies by half to encourage clients to diminish water consumption and invest in low-water-demand sectors to avoid increased prices. Also, to prevent decreasing the virtual water levels in the province, the production growth rate of some high-water-demand crops and industries is reduced to 0. Also, by raising investments in the wastewater system, this policy attempts to help preserve water resources
Focus on Water Conservation (FWC)	This policy primarily seeks to preserve water resources. The government reduces the production rate and moderates investments in high-water-demand sectors to diminish water consumption and bring the virtual water level back to the permissible range. As it imposes economic pressure on the public, this policy will cause social and political costs to the government

**Table 3. Variable Quantities in the Proposed Policies**
	Policy 1: BAU	Policy 2: FED		Policy 3: FSD	Policy 4: FWC
		Mode 1: FED1	Mode 2: FED2
The rate of closure of unauthorized wells per 10 million cubic meters of water shortage	0	0	0	2	4
Percentage profit margin for home water sales	$-$ 74%	$-$ 85%	$-$ 85%	0%	10%
Percentage profit margin of water sales in the industrial sector	$-$ 100%	$-$ 100%	$-$ 100%	$-$ 50%	$-$ 25%
Investment shift from high water demand to low water demand	0	0	0	10%	30%
Coefficient of investment attraction in industries with low water requirements	24%	24%	24%	70%	100%
Investment attraction coefficient in an industry with high water requirements	76%	76%	76%	30%	0%
Red meat production growth rate	9%	15%	15%	2%	$-$ 5%
Growth rate of white meat production	3%	10%	10%	2%	$-$ 5%
The share of investment in changing the cultivation pattern	19%	19%	19%	20%	25%
The share of investment in the elimination of agricultural products with high water requirements	0%	0%	0%	5%	15%
The share of investment in agricultural technology	25%	25%	25%	25%	25%
The share of investment in industries with low water needs	0%	0%	0%	10%	15%
The share of investment in the sewerage system	7%	0%	0%	15%	20%
Water transfer from other basins (m³)	0	0	250,000,000	0	0
Steel export growth rate	15.80%	20.00%	20.00%	0.00%	$-$ 10.00%
Ceramic tile export growth rate	6.40%	10.00%	10.00%	0.00%	$-$ 10.00%
Growth rate of fruits and vegetables	5.09%	10.00%	10.00%	0.00%	$-$ 10.00%
Egg production growth rate	$-$ 1.79%	5.00%	5.00%	0.00%	$-$ 5.00%
The growth rate of milk and dairy production	8.58%	10.00%	10.00%	0.00%	$-$ 10.00%

A total of 12 scenario-policy combinations could be derived by considering three scenarios and four policies. We applied important variables and model performance indexes to analyze the efficiency of different policies under scenarios. We should note that the model outcomes might show no significant changes after adopting specific scenario-policy combinations.

5. Results and Discussion

Once we completed the stock and flow model, we analyzed the model results. We have used the Vensim DSS Software to simulate the resulting system dynamics model and analyze its sensitivity to different variables. After model validation, we executed the model for combinations of scenarios and policies in different time horizons and analyzed the results.

5.1. Model validation

We have used several tests to validate the model, including the model behavior and the dimensional consistency tests. We used the model behavior test to ensure whether the simulation model performed correctly and whether the results were close to reality. In other words, this test evaluates the external validity of this model. In addition, we have used the dimensional consistency test to check the internal validation and the model structure. We ran the simulation model for 7 years based on the initial data of the year 2010 in the Yazd Province and compared the results with the historical data in the years 2011–2018 (Ministry of Jihad & Agriculture 2010–2020; Statistics Center of Iran 2010–2019; Yazd Management and Planning Organization 2015–2020). The results shown in Figure 9 indicate the validity of the proposed model.

Figure 9. Model Validation for Years 2011–2018

5.2. Results

As Figure 10 shows, the water supply in Yazd Province will increase in the future years in different scenarios and policies. However, this water supply increases incrementally due to (1) the gradual increase in treated municipal wastewater and (2) increased water demand in the province. We removed the results of the 50-year horizon due to the different scale of results of the scenarios FSD and FWD. However, Table 4 summarizes the values for different scenarios and policies in all horizons. The baseline and optimistic scenario results show that the FED2, which considers inter-basin water transfer, has the highest water supply in the short-term (10 years) and mid-term (20 years) horizons. However, the maximum water supply in the long term (50 years) belongs to the FWD policy. The simulation of the model for a time horizon to 2070 changes the long-term behavior of the model such that FWC will result in the highest water supply after 2052. In this case, FSD is the second priority after 2052. In the optimistic scenario, the FWD policy has the highest water supply in all horizons (see Table 4). We can conclude that the increase in inter-basin water transfer and groundwater extraction in FED1 and FED2 would not significantly affect the water supply in the long term. Accordingly, the focus on water conservation and virtual water level elevation has a better impact on water supply.

**Table 4. The Results of Model for Different Scenarios and Policies**
Variables			Water Supply (Bm³)	Water Shortage (Bm³)	Ground Water Change (Bm³)
Baseline Value			0.47	0.6	—
Pessimistic Scenario	BAU	10 Years	0.25	1.85	$-$ 2.5
		20 Years	3.1	11	$-$ 5.6
		50 Years	103E+3	$-$ 84E+3	$-$ 33.3
	FED1	10 Years	0.25	3.1	$-$ 2.2
		20 Years	0.35	19.1	$-$ 3.3
		50 Years	1.4	18.8E+3	$-$ 10.1
	FED2	10 Years	0.38	3	$-$ 2.2
		20 Years	0.43	19	$-$ 3.3
		50 Years	1.4	18.8E3	$-$ 10.1
	FSD	10 Years	0.5	0.58	$-$ 2.5
		20 Years	8.7	1.4	$-$ 3.9
		50 Years	229E+3	$-$ 246E+3	26.4E+3
	FWD	10 Years	0.91	$- 0.82$	$-$ 2.6
		20 Years	16.9	$- 12.8$	0.59
		50 Years	298E+3	$- 392$ E3	82.8E3
Baseline Scenario	BAU	10 Years	4.5	4.1	$-$ 9.5
		20 Years	60.2	53.8	$-$ 125.5
		50 Years	294.2E+3	368E+3	$-$ 1.01E+6
	FED1	10 Years	4.5	5.4	$-$ 9.4
		20 Years	60.2	59.1	$-$ 124.4
		50 Years	294.2E+3	368E+3	$-$ 1.01E+6
	FED2	10 Years	4.8	5.1	$-$ 9.5
		20 Years	60.4	58.8	$-$ 124.4
		50 Years	294.2E+3	368E+3	$-$ 1.01E+6
	FSD	10 Years	4.4	3	$-$ 8.4
		20 Years	58.4	48.1	$-$ 178.9
		50 Years	9.6E+6	$-$ 10.5E+6	$-$ 462E+3
	FWD	10 Years	3.3	1.8	$-$ 9.3
		20 Years	59.6	19.8	$-$ 169.9
		50 Years	12.3E+6	$- 16.2$ E+6	1.44E+6
Optimistic Scenario	BAU	10 Years	17.8	9.1	$-$ 18.7
		20 Years	717	313.7	$-$ 678.4
		50 Years	70.4E+6	29.8E+6	$-$ 66.5E+6
	FED1	10 Years	17.8	10.3	$-$ 18.4
		20 Years	717	319	$-$ 670.9
		50 Years	70.4E+6	29.8E+6	$-$ 66.2E+6
	FED2	10 Years	18.3	9.9	$-$ 18.5
		20 Years	717.7	318.2	$-$ 670.9
		50 Years	70.4E+6	29.8E+6	$-$ 66.2E+6
	FSD	10 Years	17.7	7.9	$-$ 18.7
		20 Years	708.1	297	$-$ 881.3
		50 Years	537.8E+6	$-$ 519.2E+6	$-$ 158E+6
	FWD	10 Years	12.8	5.6	$-$ 17.2
		20 Years	494.8	218	$-$ 812.3
		50 Years	675.9E+6	$-$ 850.7E+6	$-$ 23.8E+6

According to Figure 11, the simulation results of water shortage indicated that FWC is preferable over the other policies in all scenarios and would decrease water shortage within the province in the future. The results demonstrated an increasing trend in a water shortage for all policies in the 10 and 20 years horizon. However, the FSD and FWD policies have moderated the water shortage and changed it to a water surplus in the 50 years horizon. For the baseline scenario, the water shortage has changed from 0.6Bm³ in 2018 to a water surplus of 16.2E+6Bm³ in 50 years. As a comparison, the water shortage in the province has been predicted to be about 368,000m³ in 2070 in the business-as-usual policy (see Table 4).

Figure 12 depicts the results of the change in groundwater level. The results show that the FSD and FWD policies have the minimum reduction in the groundwater level for different scenarios and horizons. The FWD policy in the long term shows a trend change in the groundwater decrease to the groundwater increase just after 2062. The FWD policy leads to a positive groundwater change of 1.44E+6 compared with a negative amount of $-$ 1.01E+6 in the BAU policy (see Table 4).

The long-term results of FWC shows that the water supply enhances by preventing the cultivation of high-water-demand crops and importing them from the neighboring provinces. Statistics from 2011 to 2018 indicated that nearly 40% of the changes in the cultivation pattern occurred from 2011 and 2018, but a set of high-water-demand crops were still cultivated in the province (Ministry of Jihad & Agriculture 2010–2020). The simulation results of the optimistic scenario showed that BAU policy would lead to an agricultural water demand of 607 E+6m³ in the province until 2023, while Policy FWC would reduce the agricultural water demand to 377 E+6m³ by importing high-water-demand crops from the neighboring provinces rather than cultivating such crops and by raising water tariffs. The results show that while the FWC leads to lower water supply in the short run, it is preferable over the other policies in the long run as it would result in greater water supply than FED2. Although the FWC would yield fewer employment opportunities than FED in the short run, FWC would create a higher number of employment opportunities than the other policies in the long run. Since it is based on reducing the production rate and moderating investments in high-water-demand sectors, FWC brings employment opportunities in low-water-demand sectors (i.e., service sectors). Thus, FWC is preferable in terms of job creation over policies based on inter-basin water transfer and agricultural and industrial development. Moreover, the final surface water amount would be greater under FWC than under the other policies in light of increased treated wastewater. As a portion of water sales profits is invested in the wastewater sector of the province, surface water would be enhanced. The virtual water level of the province would be higher under FWC than under the other policies. The simulation results indicated that Policy BAU and the optimistic scenario would diminish the virtual water level to $- 725$ E+6m³ until 2023, whereas FWC would even raise the virtual water level to $28$ E+6m³.

The results show that the FWC leads to a higher water supply and lower water shortage for the province. In FWC, the government reduces the production rate and investments in high-water-demand sectors to decrease water consumption and bring the virtual water level back in the permissible range. The government attempts to increase water tariffs to encourage customers to reduce their consumption. The government enhances water productivity by raising water prices to motivate demand management and investing the funds in the technological development of the agricultural sector, changing the cultivation pattern, eliminating high-water-demand crops and paying compensations to farmers, developing the wastewater treatment system, and making more investments in low-water-demand industries. Moreover, the virtual water exports of the province are diminished by imposing strict regulations on the cultivation of high-water-demand crops and preventing the development of high-water-demand industries. It should be noted that FWC would cause social and political costs to the government as it brings economic difficulties to the public. Hence, it is required to moderate tariffs gradually.

5.3. Sensitivity analysis

Once FWC was found to be the best policy, sensitivity analyses were performed to measure the sensitivity of the basic variables to changes in FWC for the upper and lower bounds of the variables. The results shows that the “groundwater level” was found to be highly sensitive to “closing illegal wells” under all the scenarios. Thus, the closing of wells would strongly affect the groundwater level. The “agricultural water demand,” “treated wastewater,” and “water revenues” were found to be highly sensitive to the “domestic water sales profit margin” and “industrial water sales profit margin” under all the scenarios. Increased water prices and investing the resulting income in wastewater treatment help enhance treated wastewater in the province and reduce agricultural water demand. As a result, it is suggested that environmental bank accounts are opened to collect water sales profits and invest in the protection and revival of ecosystems by purchasing water from low-efficiency farmlands. Moreover, it is suggested to raise the prices of energy carriers and water as a long-term manner to control the industrial, agricultural, and domestic water consumption and encourage them to save water. We also suggest to provide economic incentives to industries, farmers, and domestic clients to compensate their losses in the short run.

The results also shows that the “water shortage” variable was highly sensitive to investment in changing the cultivation pattern under scenarios A and B. Hence, it is recommended to alternate cultivation patterns in Yazd Province by providing sufficient food security in Yazd based on regional resource availability and economic productivity.

The “water shortage” variable was observed to be highly sensitive to investment in new agricultural technologies under scenarios A and B. Thus, it is recommended to (1) providing financial resources to invest in enhancing agricultural technologies, such as novel irrigation systems and land and equipment renovation, (2) increasing collective agricultural management organizations, farmer participation, and collective capacities to manage water and agriculture and enhance the welfare of farmers and rural areas by granting more advantages, and (3) making investments in the agricultural sector to improve economic productivity by modernizing, industrializing, and economizing the agricultural sector.

The “water shortage” is sensitive to investment of 15% of water sales revenues in the elimination of cultivating high-water-demand crops and paying compensations to farmers in all three scenarios. Investing 25% of water sales revenues in the elimination of high-water-demand crops and 15% in changing the cultivation pattern reduced the water shortage, significantly. As can be seen in Figure 11, the curve initially followed a descending trend and then began to rise due to the dominance of other loops and factors, such as population growth and increased domestic, industrial, and agricultural water demands. However, as can be seen, different coefficients of the three variables produced different water shortage levels. This implies the high sensitivity of water shortage to the investment in the elimination of high-water-demand crops.

Figure 13. Sensitivity of Water Shortage to Investment in Eliminating High-Water-Demand Crops

The “virtual water level” showed low sensitivity to the “white meat production rate” and “red meat production rate.” Thus, animal husbandry has no significant impact on the reduction of the virtual water level as a large portion of livestock forage is imported into Yazd Province — only a small portion of the forage demand is produced within the province. Hence, it is suggested that the current husbandry rate is maintained. Moreover, the “virtual water level” had significantly high sensitivity to the “steel production rate.” Under Policy FWC and the optimistic scenario, a rise in the steel production rate from 10% to 20% would reduce the virtual water level by $31 \times 1 0^{6}$ m³ until 2023. As a result, it is necessary to largely change investments in the industrial sector by considering water limitations, regional and transregional environmental impacts, and controlling the development of high-water-demand industries in order to shift investments from such industries to low-water-demand ones. The results also show that the “virtual water level” has high sensitivity to the “ceramic tile production rate.” Under Policy FWC and the optimistic scenario, an increase in the production rate of ceramic tiles from 10% to 20% would diminish the virtual water level by 31.6 until 2023. It is also found to be strongly sensitive to the production rates of “vegetable and fruits,” “eggs,” and “dairies.” Hence, reducing the production of these products could have a significant effect on enhancing the virtual water level in Yazd Province.

6. Conclusion

Factors such as population growth and the excessive consumption of natural resources have significantly impacted the environment, including water resources. Water resources have long been a crucial aspect of the countries’ domestic politics, particularly in arid regions such as the Middle East. Yazd Province is among the aridest provinces of Iran and usually suffers from a water challenge. Since water shortage is a rough problem, measures to solve it may induce other problems that could appear after years. A non-holistic approach to the problem with no comprehensive insight could have undesirable outcomes. In this study, we adopted system dynamics as it has a holistic, systematic, and dynamic approach. We incorporated 230 influential factors and their relationships in the proposed simulation model of water resources in Yazd Province. Defining optimistic, baseline, and pessimistic scenarios with four policy packages, BAU, FED, FSD, and FWC, we executed the proposed model in the 10, 20, and 50 years horizons.

The results revealed that water transfer would not be an effective solution in the long run as it increases water shortage, even though it may reduce water shortage in the short run. Results demonstrated that FWC and the FSD policies have the best results among the proposed policies. FWC policy preserves and enhances the current water resources through investments to enhance water productivity in all areas and increase the virtual water level. It improves the water supply, water shortage, and other indicators, modifying the economic development pattern in Yazd. For the baseline scenario, the water supply has changed from 0.47Bm³ in 2018 to a water supply of 12.6E $+$ 6 in 50 years. Also, the water shortage has changed from 0.6Bm³ in 2018 to a water surplus of 16.2E $+$ 6 in 2070. As a comparison, the water supply and water shortage in 2070 in the business-as-usual policy have predicted about 294,000 and 368,000, respectively. In the FWD and even the FSD policies, we defined preventing the cultivation of high-water-demand crops, raising water tariffs, more investment in the technological development of the agricultural sector, developing the wastewater treatment system, closing illegal wells, and making more investments in low-water-demand industries. Also, it is necessary to modify the governmental management structure of water resources and enhance the authority of the Department of Environment. It is critical to separate the responsibility of water resource supply from water resource allocation.

Appendix. Important Stock and Flow Diagram Equations

Variables	Equations
Domestic Water Demand	Domestic water consumption $intensity * Domestic$ GDP
Agricultural Water Demand	Initial demand for agricultural water – decreased amount of water consumption via eliminating water-required crops
IMS Water Demand	IMS water consumption $intensity * IMS$ GDP
Consumption of Developing Funds	available $funds * Developing$ Funds consumption Rate
Domestic Water Profit	IF THEN ELSE [Profit margin of domestic $Water < 0$ , 0 (Marginal Cost for domestic $Water * Domestic$ water $demand * (1 + Profit$ margin of domestic Water))]
IMS Water Profit	(Marginal Cost for IMS $W a t e r * IMS$ water $demand * (1 + Profit$ margin of IMS Water))
Rainfall	(RANDOM NORMAL (28.3, 152.2, 97.99, 83.8, $0) ∕ 1000) * Area$ of Yazd Province
Total Water Demand	Agricultural water $demand + domestic$ water $demand + IMS$ water demand
Water Shortage	Total water demand – total water supply
Initial Inter-Basin Water Transfer	Initial water outlet from Yazd $basin + initial$ Water inlet from other $Basin + Initial$ Inlet water from the Kowsar basin
Virtual Water Level	Agricultural virtual water $import ∕ export + IMS$ virtual water $import ∕ export + Livestock$ and poultry virtual water import/export
Final Volume of Surface Water	Running $water + volume$ of treated wastewater
Running water	$Rainfall * (1 - Evaporation rate - Soil$ penetration coefficient)
Amount of Precipitation Absorbed in the Aquifers	$Rainfall * Soil$ penetration coefficient
Renewable Water Volume	final volume of surface $water + Normally$ removable groundwater
Reducing the Volume of Groundwater	Initial groundwater $extraction * (1 + Percentage$ of water harvest $growth) - Reduce$ water extraction by closing wells
Increasing the Volume of Groundwater	Absorption volume of $wastewater + amount$ of precipitation absorbed in the aquifers
Absorption Volume of Wastewater	(Agricultural water $demand * Percentage$ of agricultural water absorbed in $groundwater) + (The$ volume of wastewater produced — The volume of treated wastewater)
Volume of Treated Wastewater	Investment in wastewater treatment $system * Effect$ of investment in wastewater treatment system on the volume of treated wastewater

References

Ahmad, S and SP Simonovic [2004] Spatial system dynamics: New approach for simulation of water resources systems. Journal of Computing in Civil Engineering, 18(4), 331–340. Crossref, Google Scholar
Bagheri, A and SA Hosseini [2011] A system dynamics approach to assess water resources development scheme in the Mashhad plain, Iran, versus sustainability. ASCE Conference on the 4th International Perspective on Water Resources & the Environment, 4-6 January, Singapore. Google Scholar
Balali, H and D Viaggi [2015] Applying a system dynamics approach for modeling groundwater dynamics to depletion under different economical and climate change scenarios. Water (Switzerland), 7(10), 5258–5271. Google Scholar
Bawary, SA, M Karami-Dehkordi and G Layani [2022] System analysis of the role of social capital in sustainable rural livelihood (case study: Villages of Qarabagh County, Ghazni Province, Afghanistan). Journal of Agricultural Economic Research, 3(14), 1–22. Google Scholar
Dastkhan, H and MS Owlia [2014] What are the right policies for electricity supply in Middle East? A regional dynamic integrated electricity model for the Province of Yazd in Iran. Renewable and Sustainable Energy Reviews, 33, 254–267. Crossref, Google Scholar
Gohari, A, A Mirchi and K Madani [2017] System dynamics evaluation of climate change adaptation strategies for water resources management in Central Iran. Water Resources Management, 31(5), 1413–1434. Crossref, Google Scholar
Gohari, A, S Eslamian, A Mirchi, J Abedi-Koupaei, A Massah Bavani and K Madani [2013] Water transfer as a solution to water shortage: A fix that can backfire. Journal of Hydrology, 491, 23–29. Crossref, Google Scholar
Langarudi, S [2019] Does socioeconomic feedback matter for water models? Ecological Economics, 159, 35–45. Crossref, Google Scholar
Madani, K (2007). Water transfer and watershed development: A system dynamics approach. Proceeding of the 2007 World Environmental and Water Resources Congress, Tampa, Florida, Kabbes, KC (ed.), ASCE, 10.1061/40927(243)551. Google Scholar
Mirchi, A, K Madani, D Watkins and S Ahmad [2012] Synthesis of system dynamics conceptualization of water resources problems. Water Resources Management, 26, 2421–2442. Crossref, Google Scholar
Mehr Azar, A, A Masah Bovani, M Mashal and H Rahimi Khob [2016] Integrated modeling of water resources, agricultural and socio-economic systems of Hashtgerd plain with dynamic systems approach. Water and Irrigation Management, 6(2), 263–280. Google Scholar
Ministry of Jihad & Agriculture [2010–2020] Statistics of agricultural and horticultural products of Yazd Province, Yazd, Iran. Google Scholar
Mokhtar, A and S Aram [2017] Systemic insights into agricultural groundwater management: Case of Firuzabad Plain, Iran. Water Policy, 19(5), 867–885. Crossref, Google Scholar
Owlia, MS and H Dastkhan (2012). An application of system dynamics in electricity supply systems: Case of Yazd regional electricity company in Iran. In 30th International Conference of the System Dynamics Society. St. Gallen, Switzerland. Google Scholar
Shannak, S, D Mabrey and M Vittorio [2018] Moving from theory to practice in the water–energy–food nexus: An evaluation of existing models and frameworks. Water-Energy Nexus, 1(1), 117–125. Crossref, Google Scholar
Statistics Center of Iran (2010–2019). Statistical yearbook of Yazd Province, Yazd, Iran. Google Scholar
Stave, KA [2003] System dynamics model to facilitate public understanding of water management options in Las Vegas, Nevada. Journal of Environmental Management, 67, 303–313. Crossref, Google Scholar
Sterman, J [2000] Business Dynamics. McGraw-Hill, Inc. McGraw-Hill Education, MA, USA. Google Scholar
Water Thinking Center of Yazd University (2020). Statement No. 2 of the Yazd water perspective, Yazd University. Google Scholar
Xi, X and KL Poh [2013] Using system dynamics for sustainable water resources management in Singapore. Procedia Computer Science, 16(6), 157–166. Crossref, Google Scholar
Xu, ZX, K Takeuchi and H Ishidaira [2002] Sustainability analysis for Yellow River water resources using the system dynamics approach. Water Resources Management, 16, 239–261. Crossref, Google Scholar
Yang, CC, LC Chang and CC Ho [2008] Application of system dynamics with impact analysis to solve the problem of water shortages in Taiwan. Water Resources Management, 22, 1561–1577. Crossref, Google Scholar
Yang, J, K Lei, W Meng and S Khu [2014] Assessment of water resources carrying capacity for sustainable development based on a system dynamics model: A case study of Tieling City, China. Water Resources Management, 29(3), 885–899. Crossref, Google Scholar
Yazd Management and Planning Organization (2015–2020). Selection of economic, social and cultural indicators and indicators of Yazd province. Google Scholar
Zomorodian, M, S Lai, M Homayounfar, S Ibrahim, S Fatemi and A El-Shafie [2018] The state-of-the-art system dynamics application in integrated water resources modeling. Journal of Environmental Management, 227, 294–304. Crossref, Google Scholar