Introduction

Historically, water resource management has been directly linked to economic development and human activities. In the past century, every continent has experienced water scarcity issues, and global water use has increased two times faster than the rate of population increase1. Seckler et al.2 studied approaches in 118 countries from 1990 to 2025 and concluded that a quarter of the global population and a third of the population in developing countries will experience severe water scarcity issues in the next century. In particular, people in arid and semiarid regions are more vulnerable to water scarcity problems. Recent studies have reported an increase in drought severity in Afghanistan and its neighbouring countries in Central Asia, including Iran3,4,5,6. Hence, the proper management of water resources is important for reducing vulnerability and improving economic development and the human standard of living.

Afghanistan, as a landlocked country, highly depends on the agricultural sector. A Afghanistan National Statistics and Information Authority report from 2018 noted that the agricultural sector contributed to 18.6% of the country’s total gross domestic product (GDP) in 20187. Regardless of GDP, approximately 80% of the Afghan population lives in rural areas, and their livelihood is directly or indirectly related to agricultural activities8. The water sources for irrigation are mainly surface water and groundwater, and 86% of the irrigation water supply comes from surface water9. The irrigation water distribution uses an informal traditional system for generation. The operation of the traditional irrigation system is implemented by water users headed by mirabs10. Mirab, a combined Arabic and Persian word, translates as watermaster, and this term is also used in Iran and Central Asia. Mirabs are selected by local farmers and paid a specific amount of wheat for their water allocation and canal maintenance services. They typically spend much time walking the canal system, checking or monitoring the water allocation regime and assigning labour forces from the local community for system maintenance. The mirab system is an old tradition that has remained for centuries and survived the period of civil war in Afghanistan. The mirabs control the main, secondary, and tertiary canals, but they do not have support from or the authority of higher institutions, such as village and provincial councils or the national government, for water allocation to canals from surface waterbodies11.

The agricultural sector in Afghanistan has experienced degradation in the past decades of war and conflict due to the collapse of irrigation infrastructure. As reported by Mahmoodi8, Afghanistan has \(7.9 \times 10^{6} \;{\text{ha}}\) of arable land, and in 1980, \(3.3 \times 10^{6} \;{\text{ha}}\) of land was cultivated; however, just \(1.8 \times 10^{6} \;{\text{ha}}\) is currently cultivated due to the collapse of portions of the irrigation infrastructure. Currently, uncertainty in water allocation is another issue in Afghanistan irrigation water management. Upstream water users consume more water than those in downstream areas, and downstream farmers receive a less water than upstream farmers, especially during low-flow years12. The current irrigation challenges in Afghanistan can be categorized into two major issues: (1) lack of the proper structure and (2) lack of proper management practices for water resources. To overcome the current problems, Afghanistan needs hard and soft approaches. Hard approaches include the rehabilitation of the existing infrastructure and the construction of new irrigation structures with improved efficiency. Soft approaches refer to policy development for water allocation and management among water users considering location differences in the basin.

As reported by the FAO, in 2016, the number of chronically undernourished people in the world was estimated to have increased to 815 million, up from 777 million in 201513. As stated by Dinar et al.14, the food security situation in sub-Saharan Africa, southeastern Asia, and western Asia has worsened, especially in locations that have experienced armed conflict or natural disasters. With the rapid growth of the food demand, water scarcity and natural resource issues have become major global policy concerns. Long-term water resource management requires the capacity to evaluate different scenarios of future conditions considering uncertainty and risk.

The Soil and Water Assessment Tool (SWAT) is a useful tool for performing irrigation and chemical yield assessments in watersheds. Scholars have used the SWAT model to investigate different possible scenarios in irrigation water management and agricultural practice worldwide. For instance, SWAT was utilized to predict the impact of alternative agricultural management practices on water quality and quantity in the State of Saxony, central Germany15. In the mentioned work the authors performed a SWAT-based parameter sensitivity analysis to improve the modelling scale, and a model was used to analyses the effects of different management practices (e.g., conventional tillage, conservation tillage, and no tillage) on different crop types. Another study utilized the SWAT model and genetic algorithms to optimize the reservoir operation over two reservoirs in Gang River Basin India16. In this study, the authors considered multiple scenarios to optimize the dam operation to satisfy the irrigation water consumption in addition to power generation. Panagopoulos et al.17 used SWAT to assess the cost effectiveness (CE) of irrigation water management practices in catchments under water stress in Pinios, Greece, based on six best management practices (BMPs). The authors investigated the CEs of deficit irrigation, precision agriculture, wastewater reuse, conveyance efficiency improvements and combinations of these factors to evaluate the management practices in the study area in the context of water scarcity17.

Continuous monitoring of drainage water quantity and quality is essential for better understanding hydrological dynamics in intensely irrigated watersheds18. However, long-term data collection is expensive and time consuming19. On the other hand, generalization of site-specific experimental results to the regional level in a complex watershed with mixed soil and land use is difficult19. In this context using watershed modelling tools and remote sensing datasets can be useful for testing management practices at the watershed level. Traditional water practices are still applied for irrigation water management in developing countries and are accepted by local farmers. The local government in those countries prefers not to change the water management to avoid water conflict between users. Considering the population growth and excessive needs of water consumption the traditional water practice need to be replaced by scientific approaches. Hence, the objective of this article is to utilize watershed computer models to test the best management practice (BMPs) on large watersheds with low dataset availability. In undeveloped watersheds, hydrological models can identify the capacities and weaknesses of water resources prior to detailed study and reduce the costs for further investigations and planning. The findings of this study can also contribute to policy and decision-making processes considering the hydrological conditions of basins and other factors.

Materials

Study area

Afghanistan, which has an arid to semiarid climate, is divided into five major basins: the Northern, Harirod Murghab, Helmand, Kabul (Indus), and Amu Darya River basins (Fig. 1a). Furthermore, the NRB is divided into four river basins: Khulm, Balkhab, Sar-i-Pul, and Shirin Tagab. This study was conducted in the BRB (Fig. 1b), which is part of the NRB. The average annual precipitation in the BRB from 2010 to 2018 was 247 mm, and most precipitation occurs in snow form during the winter and as rainfall in the spring45. The Balkhab River starts from a series of six lakes in the southern part of the basin and flows through flat lands in the north. Five hydrometeorological stations observe climate data on a daily time scale in the BRB.

Figure 1
figure 1

Map of the study area; (a) Afghanistan map showing the five major basins and elevation, (b) Balkhab River basin (BRB) with the Balkhab River, five hydrometeorological stations, and eleven irrigation canals downstream. (ArcGIS Desktop, Arcmap version10.6 Drawn by author, www.esri.com).

Full size image

The agricultural sector consumes most of the surface water in the BRB to irrigate 296,744 ha of arable land. The irrigation water is distributed through 101 traditional and engineered irrigation canals and 3 small regulator dams located downstream. All available water in the Balkhab River is consumed within the basin or evaporates in the irrigation canals prior to reaching the Afghanistan border. The downstream portion of the BRB is facing a water shortage in the irrigation sector, and the irrigation canals downstream do not receive water or receive inadequate quantities of water for irrigation in six months of the year. Based on the records at the Rabat-i-Bala station on the Balkhab River, the mean annual water volume from 2010 to 2018 was \(1657 \times 10^{6} \;{\text{m}}^{3}\), and a considerable amount of flow comes from floods in the early spring season. The water transported during the flood seasons cannot be used since there is no controlling hydraulic structure to store the water for later use.

Datasets

In this study, SWAT was adopted to develop a hydrological model of the study region and simulate surface runoff. The input datasets used for the model setup were as follows: maximum and minimum temperatures, relative humidity, precipitation, solar radiation, wind speed, a digital elevation model (DEM), land cover, soil type, monthly average discharge, and irrigation canal discharge (Table 1).

Table 1 Summary of the datasets used in the SWAT modelling.
Full size table

Daily climatic data for the maximum and minimum temperatures, relative humidity, precipitation, and river discharge were obtained from the five hydrometeorological stations in the BRB (Fig. 1b). The wind speed and solar radiation datasets were retrieved from the NASA Prediction of World Energy Resources at the same geographical locations as the five stations. The climate datasets from 2010 to 2018 are depicted in Fig. 2.

Figure 2
figure 2

Average daily discharge, temperature, solar radiation, wind speed, precipitation, and snow cover area (%) datasets from 2010 to 2018 at five hydrometeorological stations: (a) Rabat-i-Bala, (b) Pul-i-Baraq, (c) Delmarogh, (d) Doshqadam, and (e) Nazdik-i-Nayak stations. (R version R- 4.1.2, drawn by author).

Full size image

DEMs with a spatial resolution of 12.5 m were obtained from the ALOS PALASAR Radiometric Terrain Correction (RT1) dataset provided by the Alaska Satellite Facility Distributed Archive Data Center (ASF DAAC). RT1 products are generated from high-resolution and moderate-resolution DEMs20. Then, the DEMs were used to create a slope map in the SWAT model.

The land cover map used was retrieved from the Agricultural Market Information Services System of the FAO in collaboration with the Ministry of Agriculture, Irrigation, and Livestock, Islamic Republic of Afghanistan. The land cover map was created using moderate-resolution satellite images from SPOT-4 and the Global Land Survey (GLS) Landsat Thematic Mapper, high-resolution satellite imagery, aerial photographs and ancillary data21. Based on the land cover map, the BRB land use is dominated by range land and grassland, with 62.25% coverage; agricultural land covers 23.12% of the basin, and the remaining areas include urban land, barren land, forest, sand/desert, and water bodies.

In many parts of the world, a proper and reliable soil type map does not exist, but the global soil type maps of FAO/UNESCO and the Harmonized World Soil Database (HWSD_v121) can be used instead22. The FAO/UNESCO soil map of the world and corresponding data were retrieved and used in this study. This soil map was developed from a statistical analysis of 4353 soil profiles from the World Inventory of Soil Emissions (WISE) database, which was developed at the International Soil References and Information Centre (ISRIC) to provide geographic descriptions of soil factors that influence global change23. Further descriptions and information regarding the FAO/UNESCO soil map of the world were provided by Doe24.

A large portion of the available water in the Balkhab River is used for agricultural water consumption. The water distributed to the canals is handled by the Northern River Basin Authority (NRBA). The amount of water discharged into the canals was obtained from an official report (NRBA25, pp. 24–26). The geographical locations of canal intakes and water rights datasets were obtained for irrigation water management from the report and included in the SWAT model.

Seasonal crop cultivation

The economy in Afghanistan highly depends on the agricultural sector, and 80% of the population lives in rural areas engaged in agricultural activities8. As noted by the FAO26, the crop types in northern Afghanistan include wheat, barley, rice, maize, pulses, potato, oil seed, clover, lucerne, vegetables, cotton, vegetables, melons, and opium poppy. Proper understanding of crop calendar information is essential for irrigation engineers to achieve synchronization between water delivery patterns and crop life cycles27. Such synchronization can contribute to efficient water usage for irrigation. The crop water requirements based on the crop calendar must be considered in irrigation water management and planning. Studies of Afghanistan are limited by data availability in many fields, and irrigation-related data are particularly scarce. In this study, crop water requirements were adopted from a previous study of Khost Province in southern Afghanistan28. Wali et al.28 used satellite remote sensing data to estimate crop water requirements. The mean annual temperature in Khost Province is 10.4 °C, and the mean annual precipitation is 188 mm; these values are similar to those in the BRB. Figure 3 illustrates the crop calendar based on a field survey and an interview of farmers in 2017 by Wali et al.28. The crop water requirements in different months are summarized in Table S1 in the Supplementary material.

Figure 3
figure 3

Typical cultivation season by crop type in Khost, Afghanistan, in 201728. The coloured bars indicate the monthly crop water requirements (mm). (R version R- 4.1.2, drawn by author).

Full size image

Methodology

Current irrigation status

The amount of water allocated from the main river to the main irrigation canals is decided by the governmen. The data for water allocation in the main canals are collected by the government and are used in the current study25. However, there is an exception for water allocation to downstream canals due to water scarcity issues during the year. Water allocation amounts for the 11 downstream irrigation canals (Table 2) are based on the water availability in the main river. The available water in the main river is distributed based on tax payments and the agricultural land area in different regions for the 11 canals (G. Sakhi, personal communication, November 2, 2020). The water rights for the 11 canals are summarized in Table 2.

Table 2 Irrigation land area and water rights for the 11 downstream irrigation canals25.
Full size table

Irrigation scenario

In this study, a new irrigation scenario for the BRB is defined based on the agricultural land area and crop water requirements. The crop type and cultivated land area were determined based on the national agriculture profile from 2016 to 2018. Per the NSIA7, the main cultivated crops (and associated cultivation areas) are wheat (60.0%), rice (3.3%), barley (3.3%), maize (3.6%), pulses (2.3%), vegetables (4.3%), potato (1%), onion (0.3%), fruits (9.3%), almond (0.6%), oil seed (2.6%), and others (9.4%). The crop types (and areas) considered in different agricultural land scenarios are wheat (65%), rice (5%), maize (5%), pluses (4%), spring vegetables (5%), summer vegetables (4%), and orchards (12%). In addition, based on the crop water requirements for different crop types (Table S1) and the specified irrigation area for each crop (Table 2), the monthly water requirement for 1 ha of agricultural land is estimated as shown in Table 3.

Table 3 Estimated monthly crop water requirement for 1 ha of agricultural land in the downstream BRB area.
Full size table

A substantial portion of usable water will be lost due to seepage in the irrigation canals29. Open canal water losses in Afghanistan irrigation systems are approximately 20%30,31. The water losses in irrigation canals placed in natural soil and fine or coarse sediment can range between 20–50%29. In this study, the approximate 20% of water losses in the irrigation canals is added to the water allocated for all 11 canals. Considering the data in Tables 2 and 3, as well as the 20% water losses in the irrigation canals the monthly water allocations for the 11 canals were updated based on the new irrigation scenario in the BRB, as shown in Table 4.

Table 4 Monthly required water allocations for the 11 downstream irrigation canals based on the monthly crop water requirements and 20% water losses in the irrigation canals.
Full size table

SWAT model setup with a proposed dam

The SWAT model is a physical process-based semi-distributed hydrological model that was developed to predict the impacts of water management, sediment, and the agricultural chemical yield in large watersheds at the daily scale32,33,34. The model simulates the hydrological process at daily, monthly, or annual time scales based on the daily temperature, precipitation, relative humidity, wind speed and solar radiation. SWAT is a widely accepted tool for efficient water resource assessment35 and evaluations of nonpoint source pollution and climate change in relation to water supplies and watersheds36,37. The model predicts surface runoff based on a water balance equation (Eq. 1).

$$SW_{t} = SW + \mathop \sum \limits_{t = 1}^{t} \left( {R_{i} - Q_{i} - ET_{i} - P_{i} - QR_{i} } \right)$$
(1)

where \(SW_{t}\) and \(SW\) are the respective final and initial soil water contents at time \(t\), respectively; \(R\), \(Q\), \(ET\), \(P\), and \(QR\) are precipitation, runoff, evapotranspiration, percolation, and the return flow, respectively, on day \(i\); and all units are mm of H2O.

For reproducibility, the major procedure used for SWAT model setup is summarized in Fig. 4.

Figure 4
figure 4

Methodological flow chart for hydrological modelling with SWAT. (PowerPoint, drawn by author).

Full size image

Step 1 Watershed delineation: The SWAT in the ArcGIS interface was used to simulate streamflow and divide the study area into subbasins. In this step, the study area was divided into 31 subbasins, point source data were added to the model, and a dam in subbasin 10 was defined and added. To determine the dam location, three criteria were considered: (1) the slopes of the river and streambed, with a maximum of 5% and minimum of 2% (0.02 < slope < 0.05)38, (2) the river discharge, and (3) the horizontal distance from the river within 500 m. Analyses were performed, and subbasin 10 was determined to be the most appropriate site for potential dam construction. However, a more detailed and complex approach is required for final decisions based on socioeconomic criteria in future work. The monthly flow downstream of the dam was estimated based on the monthly required water allocations, as shown in Table 4. The irrigation water consumption from September to December is relatively low to meet the requirements of the downstream environment and water supply; thus, we assume the minimum discharge downstream of the dam during this period at \(20\;{\text{m}}^{3} {\text{/s}}\).

Step 2 Defining hydrological response units (HRUs): Land cover, soil type, and slope maps of the study area were used for further division of 31 subbasins into 1532 smaller units.

Step 3 Creating input tables: The meteorological data from the five stations from 2010 to 2018 were defined and uploaded to the model database to simulate surface flow in this step.

Step 4 Calibration and validation: Parameter optimization was performed for the SWAT model using the SWAT Calibration and Uncertainty Procedure (SWAT-CUP) v5.2.139 and the sequential uncertainty fitting algorithm (SUFI2) for model calibration and uncertainty analysis. The SUFI2 details are presented by Homan et al.40 and Abbaspour39. The SWAT model was run for the BRB with a three-year warm-up period from January 2010 to December 2012. Then, the model was calibrated with data from January 2013 to December 2015 based on the monthly discharge recorded at four stations, namely, Rabat-i-Bala, Pul-i-Baraq, Doshqadam, and Nazdik-i-Nayak, located in subbasins 7, 10, 28, and 30, respectively. Next, the model was validated for the period from January 2016 to September 2018 with the remaining data. The calibrated parameters were adopted from a previous study in the BRB, and for more information on the detailed calibration procedure and optimized parameter values, please refer to the study by Hussainzada and Lee41.

Results

SWAT model calibration and validation

As described, the study period was divided into warmup, calibration, and validation periods, and model calibration was conducted using SWAT-CUP. First, the model was calibrated for the elevation parameters; then, calibration was performed for snow parameters and the seven most sensitive parameters. To address the downstream flow during the dry season from July to September, the groundwater baseflow was estimated based on point-source station-recorded discharge at 190 springs in the BRB. With the estimated baseflow, the SWAT results were markedly improved. The hydrographs in Fig. 5 show the comparison between the simulated and observed monthly discharge values at the four stations in subbasins 7, 10, 28, and 30 (Fig. 1b). The model performance was evaluated using the coefficient of determination (R2), Nash–Sutcliffe efficiency (NSE)42,45, and percent BIAS (PBAIS)43 by comparing the simulated and observed values at the four stations along the river. The simulation results exhibit good agreement with the observed monthly discharges at the stations, and the statistics for model performance are summarized in Table 5. As stated by Moriasi et al.44 the model’s performance during the calibration period is categorized as satisfactory in subbasin 28, good in subbasin 7, and very good in subbasins 10 and 30 when considering R2 values; in terms of NSE, the model performance is not satisfactory in subbasin 28, satisfactory in subbasins 7 and 30, and very good in subbasin 10; and the model performance is satisfactory at all stations except subbasin 10, which shows a very good performance. While the model performance during the validation period is satisfactory in all subbasins except subbasin 30, which is good considering R2, the model performance is satisfactory in subbasins 7 and 10 and unsatisfactory in subbasins 28 and 30 considering NSE, and the PBIAS model performed very well in subbasin 10, good in subbasin 7 and satisfactory in subbasins 28 and 30.

Figure 5
figure 5

Comparisons between the simulated and observed discharge values in (a) subbasin 7, (b) subbasin 10, (c) subbasin 28, and (d) subbasin 30 during calibration and validation periods. (R version R- 4.1.2, drawn by author).

Full size image
Table 5 Summary of statistical indicators for model performance.
Full size table

Irrigation scenario with a proposed dam

As described, the current water allocations for irrigation canals are determined by the government based on taxes, the agricultural land area and the availability of water in the main river. Available water in the river is distributed to the irrigation canals to meet their needs for the entire year. In the current study, a dam is proposed and added to subbasin 10 as a water storage structure to balance the water flows for irrigation purposes throughout the year. The current water allocation amounts for the 11 irrigation canals were estimated based on the water rights for the canals (Table 2) and the average monthly available river discharge. The available average monthly discharge for the river was retrieved from the SWAT model results for subbasin 10, and the results are summarized in Table 6. The Balkhab River experiences the highest flow in May (113 m3/s) due to rapid snowmelt and rainfall41. The driest month in the basin is September, with an average total monthly flow of 31.97 m3/s.

Table 6 Current monthly water allocations based on the average simulated discharge (2013–2018) for the 11 irrigation canals in the BRB.
Full size table

A comparison between the allocated amounts of irrigation water in the newly developed scenario (Table 4) and those based on existing practices (Table 6) indicates that the current irrigation water distribution is unreliable and cannot meet the water requirements for crop irrigation. The farmers face water stress issues during March, April, July, and August, and surplus river water is available in the other eight months. The greatest water shortage occurs in March and April, when more than \(369.9 \times 10^{6} \;{\text{m}}^{3}\) of water is needed for irrigation. The annual average flow in subbasin 10 was estimated as \(1.578 \times 10^{9} \;{\text{m}}^{3}\) by the SWAT model from 2013 to 2018. Additionally, the irrigation water requirement in the BRB was estimated to be \(1.197 \times 10^{9} \;{\text{m}}^{3}\). The Balkhab River transports enough water for the irrigation of existing agricultural lands and provides \(381 \times 10^{6} \;{\text{m}}^{3}\) of surplus water each year. The water shortages come from the timing differences between the peak irrigation water demand and the peak river flow times.

The proposed dam in subbasin 10 and the corresponding management scheme were added to the SWAT model. The hydrographs in Fig. 6 depict the river discharge in subbasin 7 before and after dam inclusion in the model. The resulting hydrographs reflect the water shortage during March and April from 2013 to 2018, and the basin experiences a small water shortage during July and August each year. As depicted in the hydrographs, May, when peak flows and floods occur, displays the highest potential for water storage. The Balkhab River has the potential to irrigate existing agricultural lands and promote further land development in the case of storing excess water during the eight months when irrigation water consumption is less than the available amount of water resources.

Figure 6
figure 6

Resulting hydrographs from subbasin 7 without (blue) and with (red) dam inclusion in subbasin 10 with 20% irrigation water losses. The grey line indicates the result without the 20% losses in the modelling. (R version R- 4.1.2, drawn by author).

Full size image

Discussion

Effects of the dam reservoir

Irrigation water consumption is highest during April in the BRB, and peak river discharge occurs in May. Additionally, river discharge cannot provide enough water in March, April, July, and August in the BRB. The Balkhab River transports enough water to irrigate the existing agricultural lands and promote the further development of agricultural land. During May, river discharge is increased by floods on certain days, and farmers cannot use these water resources. To overcome these challenges, a controlling hydraulic structure is needed. A dam control structure can store flood water for later consumption and balance the water supply and consumption demands during months with low river discharge.

In the case of dam construction in the BRB, there is potential to store water during the winter season, when agricultural water consumption reaches a minimum, and in May, when floods occur. The stored water can be managed and released based on the crop water requirements in each month to prevent the loss of agricultural products before harvesting. Such a scheme could not only increase the security of the agricultural sector and products but can also contribute to the economic development of the country and improve food security.

Referring to the bar charts shown in Fig. 7, especially the final bar charts that show the total water allocation, it is clearly illustrated that the Balkhab River not only has the potential to irrigate the current arable land areas and newly developed agricultural lands. Based on the total water allocation amounts presented in Fig. 7, the discharge in the mainstream is higher than the amount of water required for consumption in the agricultural sector during January, February, May, June, September, October, November and December and lower than the required amount in other months.

Figure 7
figure 7

Comparisons of the current and planned water distributions based on water allocations in the 11 irrigation canals and the total water allocation. (R version R- 4.1.2, drawn by author).

Full size image

Agricultural water management

The existing traditional water management is effective since water users are engaged in determining the water distribution in irrigation canals. The mirab system is acceptable by the local community and government authorities. Water allocation through the main canals by the government and the current water allocation scheme based on tax payments (water rights) and agricultural land areas should be established with a more scientific approach than is currently used. The current water allocation approach results in poor irrigation downstream and an unequal distribution of water to farmers. Subbasin 1 encompasses 88,706 ha of arable land, and the total current arable land in the BRB is 296,744 ha. The water allocated to subbasin 1 through the Aqcha canal accounts for 20.4% of all allocated water, and the arable land in the subbasin accounts for 29.8% of all arable land in the basin. The Fayaz Abad canal transmits 11.42% of the allocated river water for 17,180.5 ha of arable land. This unequal water distribution has led to ecosystem degradation and the migration of farmers to large cities in subbasin 1.

Water resource management in the BRB must be reformed to ensure an equal water distribution among users. Such a scientific approach can promote stable and sustainable development in the agricultural sector. The unequal water distribution makes downstream farmers vulnerable and increases economic and food security risks. Moreover, unequal water allocation issues, drought, and extreme weather are the main challenges for farmers in rural areas in Afghanistan. The country cannot reach its energy, irrigation, or urban and rural development goals unless it makes substantial improvements in the development and management of water resources.

In Fig. 7, the current water distribution and that based on planned water management in the BRB after the proposed dam is constructed are compared; additionally, the uneven water distribution is visualized for water users. For example, the Imam-Saib canal receives more water than is required water for irrigation in its area, and the Aqcha canal receives less water than required for its area in February, March, April, May, July, and August. The uneven water distribution among water users has caused many farmers to migrate to large cities and neighbouring countries for work. Moreover, a large portion of river discharge comes from spring floods during April and May, and the agricultural sector does not require water during the winter season, when no irrigation is necessary. Thus, based on the current water management approach in the BRB, reform is needed to improve the irrigation scheme by adjusting water allocation amounts and storing water for later use.

Agricultural land development

As mentioned in “Irrigation scenario with a proposed dam” section, the Balkhab River transports 381 million cubic metres of surplus water beyond what is needed for irrigating the existing agricultural lands. This extra water is wasted because the flows occur in the nonirrigation period in winter or during the short flood periods in the spring. Currently, 1.8 million ha of agricultural lands are cultivated in Afghanistan, and before 1980, approximately 3.3 million ha were cultivated8. Based on the data in Table 3, the water requirement of 1 ha of agricultural land in the BRB is \(4559.5\;{\text{m}}^{3}\) annually. In the case of dam construction in the BRB, the potential exists to irrigate an additional 18,470.6 ha. In the case of dam construction, the area of arable land could increase by 6.18% in the BRB. The improvement associated with dam construction could highly affect farmers’ lives and agricultural productivity. Water losses in irrigation canals are another factor that must be considered in the future development of watersheds. The outputs in Fig. 6 show that in the ideal case of zero water losses in the irrigation canals the water resource can develop up to 83,561.8 ha of new agricultural lands in the study basin.

Conclusions

In this study, the potential for irrigation in the BRB was determined to be adequate for the current agricultural lands, and an additional 18,470.6 ha of new agricultural land could also be irrigated. In the BRB, the traditional mirab system is applied for water distribution and canal maintenance. The assessments of the current policies for water management in the BRB indicate that the water distribution is unequal among farmers, and upstream agricultural lands use considerably more resources than downstream agricultural lands.

During March and April, wheat-cultivated lands, as the main cereal of the country, are at risk of water shortages in the BRB. Water scarcity issues can decrease the wheat yield or completely destroy the crop in these two months. Users cannot fully utilize the considerable potential of the water resources provided by the Balkhab River due to a lack of a control system during peak flows in the spring and winter flows; this water could be stored and used later for irrigation. The construction of controlling hydraulic structures and water resource storage during these seasons could mitigate water scarcity issues in normal years and drought periods.

The results of the current study can be utilized by policy makers to make decisions regarding water resource management in the study area. More accurate and individual studies are required for dam site selection, irrigation water allocation, and cost–benefit analysis prior to making final decisions related to dam construction. The current results can provide preliminary information for future short-term and long-term detailed studies in the BRB region in northern Afghanistan.