IWMI Research Report Managed Aquifer Recharge: The Solution for Water 151 Shortages in the Fergana Valley Akmal Karimov, Vladimir Smakhtin, Aslon Mavlonov, Vecheslav Borisov, Inna Gracheva, Fazleddin Miryusupov, Jamol Djumanov, Tatyana Khamzina, Rustam Ibragimov and Botir Abdurahmanov RESEARCH PROGRAM ON Water, Land and Ecosystems IWMI Research Report Managed Aquifer Recharge: The Solution for Water 151 Shortages in the Fergana Valley Akmal Karimov, Vladimir Smakhtin, Aslon Mavlonov, Vecheslav Borisov, Inna Gracheva, Fazleddin Miryusupov, Jamol Djumanov, Tatyana Khamzina, Rustam Ibragimov and Botir Abdurahmanov RESEARCH PROGRAM ON Water, Land and Ecosystems Research Reports The publications in this series cover a wide range of subjects—from computer modeling to experience with water user associations—and vary in content from directly applicable research to more basic studies, on which applied work ultimately depends. Some research reports are narrowly focused, analytical and detailed empirical studies; others are wide-ranging and synthetic overviews of generic problems. Although most of the reports are published by IWMI staff and their collaborators, we welcome contributions from others. Each report is reviewed internally by IWMI staff, and by external reviewers. The reports are published and distributed both in hard copy and electronically (www.iwmi.org) and where possible all data and analyses will be available as separate downloadable files. Reports may be copied freely and cited with due acknowledgment. About IWMI IWMI’s mission is to improve the management of land and water resources for food, livelihoods and the environment. In serving this mission, IWMI concentrates on the integration of policies, technologies and management systems to achieve workable solutions to real problems—practical, relevant results in the field of irrigation and water and land resources. IWMI Research Report 151 Managed Aquifer Recharge: The Solution for Water Shortages in the Fergana Valley Akmal Karimov, Vladimir Smakhtin, Aslon Mavlonov, Vecheslav Borisov, Inna Gracheva, Fazleddin Miryusupov, Jamol Djumanov, Tatyana Khamzina, Rustam Ibragimov and Botir Abdurahmanov International Water Management Institute (IWMI) P O Box 2075, Colombo, Sri Lanka i The authors: Akmal Karimov is Researcher - Hydrogeology at the Central Asia office of the International Water Management Institute (IWMI) in Tashkent, Uzbekistan; Vladimir Smakhtin is Theme Leader – Water Availability and Access at the headquarters of IWMI in Colombo, Sri Lanka; Aslon Mavlonov is Deputy Chair of the State Committee of the Republic of Uzbekistan on Geology and Mineral Resources, Tashkent, Uzbekistan; Vecheslav Borisov is Senior Hydrogeologist at the Institute of Hydrogeology and Engineering Geology in Tashkent, Uzbekistan; Inna Gracheva is a Groundwater Modeler at the Institute of Hydrogeology and Engineering Geology in Tashkent, Uzbekistan; Fazleddin Miryusupov is Chief Hydrogeologist at the Institute of Hydrogeology and Engineering Geology in Tashkent, Uzbekistan; Jamol Djumanov is a Groundwater Modeler at the Institute of Hydrogeology and Engineering Geology in Tashkent, Uzbekistan; Tatyana Khamzina is a Soil Scientist at the Design Institute Uzmeliovodhoz, Tashkent, Uzbekistan; Rustam Ibragimov is a GIS Specialist at the Design Institute Umeliovodhoz, Tashkent, Uzbekistan; and Botir Abdurahmanov is a PhD student at the Tashkent Institute of Irrigation and Melioration in Tashkent, Uzbekistan. Karimov, A.; Smakhtin, V.; Mavlonov, A.; Borisov, V.; Gracheva, I.; Miryusupov, F.; Djumanov, J.; Khamzina, T.; Ibragimov, R.; Abdurahmanov, B. 2013. Managed aquifer recharge: the solution for water shortages in the Fergana Valley. Colombo, Sri Lanka: International Water Management Institute (IWMI). 51p. (IWMI Research Report 151). doi:10.5337/2013.205 / water management / aquifers / recharge / water shortage / valleys / river basins / flow discharge / upstream / downstream / groundwater irrigation / canals / groundwater development / groundwater extraction / water storage / wells / reservoirs / artificial recharge / infiltration / irrigated land / soil profile / models / Central Asia / Fergana Valley / ISSN 1026-0862 ISBN 978-92-9090-770-1 Copyright © 2013, by IWMI. All rights reserved. IWMI encourages the use of its material provided that the organization is acknowledged and kept informed in all such instances. Front cover photograph shows Mr. Turdali Nassredinov, Consultant to IWMI, and Mr. Abdusalom Kayumov, Head of the Water Management Administration of the Besharyk District, Besharyk, Fergana Province, Uzbekistan, discussing the potential for increasing aquifer recharge (May, 2009) (Photo credit: Dr. Akmal Karimov, IWMI). Please send inquiries and comments to IWMI-Publications@cgiar.org A free copy of this publication can be downloaded at www.iwmi.org/Publications/IWMI_Research_Reports/index.aspx Acknowledgements The authors are grateful to the staff of the local water management organization of the Besharyk District of Fergana Province, Uzbekistan, for their collaboration and participation in the field studies. Special thanks are due to Mr. Tohir Aulchaev, Senior Hydrogeologist of the expedition of the HYDROENGEO Institute in the Fergana Province, Uzbekistan; Mr. Turdali Nassredinov, Consultant to IWMI, and Mr. Abdusalom Kayumov, Head of the Water Management Administration of the Besharyk District, Uzbekistan, for their valuable contributions to the field studies. The authors are also grateful to Dr. Paul Pavelic (Senior Researcher - Hydrogeology, IWMI, Southeast Asia Office, Vientiane, Lao PDR), Dr. Prathapar Sanmugam (Theme Leader – Productive Water Use, IWMI, New Delhi Office, India) and to one anonymous reviewer for their helpful comments and suggestions, which helped to improve the final version of this report. Collaborators This research study was a collaboration of the following organizations. International Water Management Institute (IWMI) The Institute of Hydrogeology and Engineering Geology (HYDROENGEO) OPEC Fund for International Development (OFID) Donors The continuous financial support provided to this study by the Organization of the Petroleum Exporting Countries (OPEC) Fund for International Development (OFID), and, more recently, by the CGIAR Research Program on Water, Land and Ecosystems (WLE) is gratefully acknowledged. OPEC Fund for International Development (OFID) This work has been undertaken as part of the CGIAR Research Program on Water, Land and Ecosystems. IWMI is a member of the CGIAR Consortium and leads this program. Contents Summary vii Introduction 1 Managed Aquifer Recharge: Concept and Examples 3 MAR in the Fergana Valley 7 The Proposed MAR Strategy for the Fergana Valley 8 Assessing MAR Potential in the Fergana Valley 10 Pilot MAR: The Isfara River Basin 14 Pilot MAR: The Sokh River Basin 26 Conclusions 40 References 41 v v Summary As a result of the growing demand for food a stepwise procedure of implementing MAR in and energy, the competition for water between the Fergana Valley, starting from the regional upstream and downstream users in the Syrdarya assessment of the MAR potential to testing MAR River Basin has increased. The change in the at the pilot scale through field and modeling upstream reservoir operation from a conjunctive studies. The regional assessment shows that over irr igation/hydropower mode to exclusively 500,000 ha, or 55% of the currently irrigated land hydropower generation resulted in reducing in the Fergana Valley, can be shifted from canal the river flow downstream in the summer and irrigation to conjunctive surface water-groundwater increasing it in the winter. This phenomenon irrigation. This will reduce the return flow to the caused a downstream water shortage of 2,000- river by 30% (or by 1,000 Mm3/year), and form 3,000 Mm3/year in the summer and an excessive, free storages of 500 Mm3 in the command areas often unutilized, flow of the same magnitude in of main canals. Pilot-scale studies for Isfara and the winter. This study suggests that the current Sokh aquifers in the Fergana Valley support practice of sequential in-channel reservoirs is the results of regional assessment. Overall, not coping well with the needs of both upstream groundwater development for irrigation and MAR and downstream water users. Furthermore, it in the Fergana Valley is expected to reduce the examines the alternative approach of managed winter flow of the Syrdarya River at the valley aquifer recharge (MAR) in the upstream of outlet by 1,500 Mm3/year, and consequently Fergana Valley with a view to adapt to new water increase its summer flow by the same magnitude. management reality. Favorable hydrogeology This report proposes a major shift in the focus conditions prevailing in the Fergana Valley are of development projects in the Fergana Valley, envisaged to create benefits from MAR both from rehabilitation of dense drainage systems to at local and regional levels. The study follows groundwater development for irrigation and MAR. vii Managed Aquifer Recharge: The Solution for Water Shortages in the Fergana Valley Akmal Karimov, Vladimir Smakhtin, Aslon Mavlonov, Vecheslav Borisov, Inna Gracheva, Fazleddin Miryusupov, Jamol Djumanov, Tatyana Khamzina, Rustam Ibragimov and Botir Abdurahmanov Introduction The Syrdarya River Basin in Central Asia with its The shift of the upstream reservoir on the main tributary – Naryn – has a catchment area of Naryn River from irrigation to hydropower 219,000 km2 and generates about one-third of the generation mode in the beginning of 1993, total flow that used to feed the Aral Sea. Irrigated and associated increase in winter discharges agriculture has been practiced in the basin from and reduced summer f low, caused an ancient times. But it was the massive scale of estimated shortage of 3 km3 of water annually flow regulation in the second half of the twentieth from the required amount for the middle century, and subsequent geopolitical changes and the downstream water requirement for in the 1990s with the formation of the newly agriculture (Mustafaev et al. 2006). The independent states that dramatically changed the coincidence of the occurrence of peaks in the hydrology of the river and complicated the overall winter hydropower releases and return flow water management of the basin downstream. from the irrigated land in the Fergana Valley Under current conditions, the middle and the results in excessive flows that complicate the downstream parts of the Syrdarya Basin face operation of the downstream reservoirs. There severe seasonal water shortages for agriculture is not enough free storage in the middle and and the environment. These shortages are the downstream reservoirs to accumulate the caused, primarily, by three factors: releases from the upstream reservoir in winter ● Non-uniform distribution of limited water. for use in the summer. The middle and the downstream of the ● Global climate change and its impact on basin generate only 10.9 km3 of flow (29% water resources. Over the last 70 years, of the long-term mean annual flow (MAF) the air temperature has increased by of the entire basin), while the needs of the 0 .029 oC per year fo l l owed by h igh downstream agriculture and environment fluctuations in precipitation. According to the are at least twice as high (Abdullaev et al. Hydrometeorology Service of Uzbekistan, 2007). As a result, with an increasing demand the reduction of the Syrdarya River flow by for water, the middle and the downstream 2050 may be around 6-10%, with increased water users become more dependent on the frequency of extreme, high and low flows, upstream inflow. which will require more storage capacity ● Growing competition for water. The growing (Agaltsceva and Pak 2007). competition for water between hydropower The cascade of reservoirs located along the generation operations located upstream, Syrdarya River, as recent history shows, is not and agriculture and environment demands able to meet the requirements of both upstream downstream also causes water shortages. and downstream users. Excessive river flow in 1 winter and lack of free storage, which causes Groundwater Modeler, Institute of Hydrogeology freshwater discharge into the saline depression and Engineering Geology, December 18, 2009 in the midstream of the river, and the resultant pers. comm.). shortage of water for irrigation in the summer, The MAR for agricultural needs in Uzbekistan are the main implications of the current water was investigated in 1970-1990 by Mirzaev, 1972; management practices in the basin. Akramov, 1991; and Sherfetdinov, 2000. During This emphasizes the need for alternative/ this period, a number of aquifers were identified additional storage capacities. One potential as having a high potential for aquifer recharge. option is associated with subsurface storage. The They were Kitab-Shahrisabz in Kashkadarya River upstream of Fergana Valley in the Syrdarya River Basin, Iskovat-Pishkaran, Osh-Aravan, and Isfara Basin has favorable hydrogeology conditions to and Sokh in the Fergana Valley (Akramov 1991). store extra winter flows for summer use. Two Free capacities of the Osh-Aravan Aquifer were main and multiple small tributaries form and estimated at 500 Mm3, and at 200 Mm3 in the Sokh feed the Syrdarya River in the Fergana Valley. Aquifer (Akramov 1991). Field studies conducted Subsurface storage, which at this stage is almost at the Kitab-Shahrisabz Aquifer demonstrated that full, is estimated to be 200 km3 (Mavlonov et al. groundwater extraction increases free storages to 2006). Renewable fraction of this water needs to 950 Mm3. Additionally, modeling of the Iskovat- be beneficially used, and this will also free some Pishkaran Aquifer indicated that groundwater storage to accommodate winter flows – using extraction with discharge of 5.5 m3/s will increase managed aquifer recharge (MAR) (Dillon 2005, free capacities to 500, 800 and 1,043 Mm3 after 10, 2011). 20 and 30 years, respectively. The main limiting First attempts to implement MAR were factor for MAR was found to be free flow of rivers exploited in Uzbekistan for municipal water rather than free storages of aquifers. In the past, supply (Mirzaev 1972; Akramov 1991). Artificial artificial groundwater recharge for agricultural groundwater recharge structures, such as open purposes in Central Asia was limited to theoretical ditches, were constructed in 1970 to store the studies and modeling. flow of the Chirchik River underground - to The main difference in agricultural water supply Tashkent City with drinking water. Similar use in the Syrdarya River Basin as compared recharge structures were used for village water to other countries of Asia is that agriculture, supply in Karakalpakistan and Khorezm in the which is entirely dependent on the canal system downstream of the Amudarya River. In the with furrow irrigation, produces a major part of sandy areas of Karakalpakistan, groundwater groundwater recharge. There are few irrigation recharge structures were constructed in several projects that are based on the conjunctive use of locations aimed to increase leakage from canals groundwater and canal water in the foothill areas during July-August (high water season), when of the Fergana Valley (Tihonova 1972; Mirzaev water salinity is low. The leakage from the 1972). Irrigation systems within Kazalinsk, Asht main canals forms temporary freshwater lenses and Dalversin project areas demonstrated benefits above the saline brackish water. This water is of conjunctive use in preventing soil salinization then extracted to supply the rural population and waterlogging. Groundwater abstraction with drinking water. Modeling studies carried achieved its maximum level in the mid-1990s, out by the Institute of the Hydrogeology and but then declined due to reduced investment Engineering Geology estimated an optimal in infrastructure in the countries undergoing regime of groundwater extraction, which would transition. facil itate the prevention of freshwater and The growing shortage of surface water in the saline water mixing, and thereby maintain the Syrdarya River Basin requires the consideration water quality in the lenses at the supply level of alternative sources including groundwater. that is acceptable for drinking purposes until Advance planned MAR activities may prevent the next high water season (Inna Gracheva, the potential negative consequences that arise 2 from a shift from canal use to conjunctive use. local and regional levels. This report reviews the Groundwater use for agriculture is low in the experiences of MAR in the arid regions of India, Central Asian region as compared to canal China, Australia and USA and proposes a way of irrigation and MAR implementation, both of which implementing MAR in the Fergana Valley. This have been adopted only to a limited extent at this report aims to bring the attention of policymakers stage. However, MAR implementation on a wide and practitioners to the benefits of adopting scale may significantly alleviate the looming water MAR practices in the region and proposes its scarcity and improve water management, both, at implementation procedure for the Fergana Valley. Managed Aquifer Recharge: Concept and Examples MAR is intended to regulate groundwater recharge ● Using groundwater extraction to increase to increase water resources, improve water leakage from riverbeds, floodplains, canals quality in subsurface horizons and regulate return and drains. flow from irrigated lands. The adoption of MAR ● Using groundwater extraction to create free practices may yield the following benefits: subsurface horizons. ● Temporarily storing (‘banking’) water in ● Effecting changes in the cropping pattern and subsurface horizons for later use. soil tillage. ● Sustaining groundwater levels and preventing In the last few decades, there has been groundwater depletion or raising the water a phenomena l inc rease in g roundwater level, minimizing salinity and waterlogging. extraction worldwide. Unsustainable groundwater ● Reducing non-processed water depletions for development was followed by drawdown of evaporation, flow to sink and pollution. groundwater levels over large areas and ● Flood control. degradation of the water quality (Shah et al. 2000). Higher rates of depletion are observed in ● Improving surface water and groundwater many countries, including India, China, USA and quality. Mexico, where increasing population pressure ● Environmental gains (for example, stored and expected economic gains resulted in the water intended for landscape irrigation or depletion of the resource (Rosegrant et al. 2002). baseflow to rivers). In some areas, this made groundwater extraction uneconomical and prompted farmers and Various methods of MAR and preparatory authorities to look for mitigation options. activities can be applied in agriculture, including the following: ● Regulating groundwater natural recharge. India ● Creating artificial groundwater recharge to India has an agriculture-based economy and increase or replenish groundwater storages. the shortfall of 174 km3 in surface water storage ● Adoption of water-saving technologies to has made groundwater resources development reduce areal or linear groundwater recharge imperative to the country. This is because surface caused by saline fluxes from the vadoze zone. water storage is needed to meet the different 3 needs of water use sectors, especially agriculture. 14% of total land area of India — is suitable The uncontrolled development of groundwater for MAR and that a volume of 36,453 Mm3 is through the construction of 19 million open wells/ available for recharge annually. These figures shallow tube wells and subsequently deep tube have been estimated in some detail at the State wells, increased the irrigated area from 6.5 level and equated to an average recharge of 80 million hectares (Mha) in 1950 to 58.5 Mha in mm over the entire recharge area. This will be 2009 (Sakthivadivel 2007; Sharma 2009). While achieved with 3.7 million rooftop structures in the area irrigated by surface water increased by urban areas and 0.225 million rooftop structures 28%, the area irrigated by groundwater increased in rural areas, 37,000 percolation tanks (each by 105% over the same period. The increased of 0.2 Mm3), 110,000 check dams (each of 0.03 development of groundwater resources therefore, Mm3), 48,000 recharge shafts/dug wells (each of met the major requirements of irrigation and 0.03 Mm3), 26,000 gully plug/gabion structures drinking water for a rural population of 700 million (each of 0.005 Mm3) and further development of and the needs of more than 50% of the urban 2,700 springs in hilly areas, among others (Central and industrial sectors. However, the unregulated Ground Water Board 2005; Romani 2005). development of groundwater in arid and semi-arid areas resulted in a continuous decline of water levels over an area of about 340,000 km2. From China 1992 to 2005, the number of ‘unsafe blocks’ (areal units) increased from 325 to 1,615, including Groundwater resources in China are unevenly 839 overexploited blocks (Kumar and Rajput distributed and utilized across the regions. The 2005). The groundwater depletion is highest annual natural recharge of fresh groundwater in western India; the number of overexploited resources in China is 884 km3, and groundwater blocks continues to grow at the present rate of resources account for about one-third of the 5.5% per annum, and it is expected that by 2018, nation’s total water resources (Ministry of Land approximately 36% of India will face serious water Resources of China 2005). About 70% of the shortages due to depletion of groundwater (Kumar groundwater resources are located in southern and Rajput 2005). China, and only about 30% are found in northern While declining groundwater levels cause China. Here, the intensity of groundwater use, huge environmental, social and economic costs, however, occurs in a much different pattern. Rural there is potential for increasing groundwater and urban users in northern China are using more recharge. The total annual precipitation is 4,000 than 70% of the known groundwater resources km3, of which about 1,240 km3 forms surface in the region. In contrast, less than 30% of the runoff. It has been estimated that 872 km3 is known groundwater resources in southern China still available for recharge and it is feasible to are being used (Jinxia et al. 2007). In the early have a subsurface storage of 214 km3 (Tuinhof 2000s, groundwater use exceeded 100 km3 or et al. 2003). Being aware of this potential, India 20% of the total water utilization of China (Ministry has developed a strong focus on groundwater of Water Resources of China and Nanjing Water recharge and is widely promoting watershed Institute 2004). However, this share is uneven, with development across India. Micro-watershed only 14% in southern China and 49% in northern management, including the construction of check China, where groundwater was, and is, critical for dams and percolation ponds, currently costs over the emergence and expansion of agriculture, in USD 500 million per year (Central Ground Water particular, and the regional economy, in general. Board 2005). A comprehensive quantification The intensive use of groundwater has also created has recently been published in the Master many environmental problems, related to an Plan for Artificial Recharge to Ground Water in overdraft in northern China, in particular (Ministry India (Central Ground Water Board 2005). It is of Water Resources of China and Nanjing Water estimated that an area of 448,760 km2 — about Institute 2004). The problem is widespread in that 4 a 48% decline was observed in the water tables overexploited shallow groundwater. The second of villages in six provinces (Wang et al. 2005). site is located in the downstream of the Yongding With a falling water table, pumping costs have River channel, which included artificial recharge risen by CNY 0.005 per cubic meter and, in many in 2001, from both shallow groundwater and deep cases, agricultural wells have been abandoned groundwater (Jia and You 2010). and replaced by new deeper tube wells (Ministry Recently, more advanced technologies were of Water Resources of China and Nanjing Water used for aquifer recharge. In 2009, well injection Institute 2004). was applied to fill the groundwater reservoir in the Two approaches were taken up to arrest the Futuo River Basin (i.e., upstream of Ziya River). problem: a) agricultural water-saving measures; From August 20 to September 7, 2009, 18 Mm3 and b) MAR. The demand management effected of water from the upstream of the Huangbizhuang through agricultural water-saving measures is Reservoir was infiltrated underground (Jia and primarily for the purpose of obtaining 50 mm/yr of You 2010). Multipurpose underground reservoirs ‘real water savings’ and thereby have a reduction were constructed in different locations in China, in groundwater abstraction for irrigation. The including: Wanghe underground reservoir in measures found to be capable of reducing the Laizhou, with a regulating storage capacity of 56.9 rate of decline in the water table are as follows: Mm3; Dagu River underground reservoir in the irrigation water distribution through low-pressure Jiaodong Peninsula with a capacity of 238 Mm3; pipes (instead of open earth canals); drip and and others (Ishida et al. 2011). The reservoirs micro-sprinkler technology; improve irrigation are built by constructing underground dams by scheduling; agronomic measures such as deep grouting or with clay walls. ploughing, straw and plastic mulching; and the use of improved strains/seeds and drought- resistant agents (Jinxia et al. 2007). Australia MAR implementation in China uses two methods: a) low-cost technologies, and b) Unmanaged aquifer recharge, or ‘intentional underground reservoirs. Low-cost technologies water-related activity known to increase aquifer include small gulley dams, diversion canals, recharge, which usually has been undertaken rubber-dams, village pits and ponds, flooding of to dispose of water rather than to recover it’ maize fields (following wet season storms), and (EPA 2009), has a long history in many cities diversion of river flow to flood-retention reserve and towns of Australia. Disposal of water in land. In the North China Plain, Xu et al. (2009) the form of roof runoff infiltration (since 1829) have identified seven specific regions that could or storm water drainage wells have been used be targeted for MAR using low-cost technologies, since the 1880s. However, the role of drainage all of which are alluvial fans in the piedmont wells in sustaining groundwater supplies was of the Taihang Mountains, where regional appreciated only much later, and steps have been recharge occurs. The source of water diverted successively introduced since the 1970s to protect for recharge could be a combination of treated groundwater quality. In the 1960s and 1970s, urban wastewater and, potentially, excess surface the significant MAR schemes in Australia were water (e.g., from southern China, delivered via the surface infiltration schemes primarily related to South to North Water Transfer Scheme) during agriculture (Charlesworth et al. 2002). Since 1990, wet years. Artificial recharge experiments were water injection and recovery from the same well, implemented in some parts of the North China called ‘aquifer storage and recovery’ (ASR), is the Plain. The first site is located in the downstream most common type of MAR employed in Australia of the Chaobai River channel; nine weirs (width: (Parsons et al. 2012). A few examples of MAR 300-400 m, height: 3-5 m) were constructed use in agriculture are given below. from 1984 to 1998 to capture releases from the Intensive groundwater extraction during 1970- upstream of the Miyu Reservoir and recharge the 1990 caused drawdown and increase in salinity in 5 many parts of southern Australia. These regions artificial recharge projects in Arizona, i.e., Salt are Angas Bremer and Barossa Valley, southern River, Central Arizona Project and others. Australia, and the Lower Burdekin, Queensland Since then, direct surface and direct subsurface (GHD Pty Ltd. and AGT Pty Ltd. 2011). The recharge methods have been successfully Angas Bremer Region is an important premier used to store water in many aquifers of the wine district in South Australia, with about 80% Arizona State. Water spreading methods using of current irrigation dedicated to vineyards. in-channel and off-channel basins are used to Extraction of groundwater in the years before store large volumes of surplus surface water. 1980 (up to about 20 Mm3/year) caused regional In 1986, the Salt River Project (SRP), which is drawdown and an increase in salinity due to Phoenix’s largest water purveyor in partnership lateral movement of more saline water from the with six municipalities of the metropolitan area basin margins and downward leakage of saline constructed and operated the State’s largest water from the overlying aquifer. From the 1980s, underground storage facility. The Granite Reef irrigators began experimenting with diverting flows Underground Storage Project (GRUSP) is a from the Angas and Bremer rivers into irrigation surface water spreading operation located in bores. The peak of MAR activities occurred in the east of Phoenix, consisting of seven basins the wet spring of 1992 when 2.4 Mm3 of water and occupying an area of 150 ha. It is built in a was injected into about 30 wells. Water pressures secondary dry channel of the Salt River, isolated in the region were measured and found to have from normal river flows, approximately 5 km risen close to levels that had not been seen since downstream from SRP’s Granite Reef Diversion the 1950s (GHD Pty Ltd. and AGT Pty Ltd. 2011). Dam. This facility has a capacity to store 250 Another example is the wine region in the Mm3/year in the aquifer. It recharges imported Barossa Valley, where water resources are water from the Salt, Verde and Colorado rivers managed according to the ‘Water Allocation Plan’. and a very small volume of reclaimed water. Some wine growers had insufficient groundwater Since 1994, this project has stored in the aquifer allocated to them to irrigate the additional area, in excess of 1,200 Mm3 of water. In 2007, the and were not able to get their licensed allocation SRP was completed and began operating the of groundwater increased. Hence, they turned to New River-Agua Fria Underground Storage MAR. At that stage, surface water resources were Project (NAUSP). This facility also uses surface not prescribed and could be accessed without a basins for recharge and has an annual storage license. Water of low salinity and turbidity was capacity of 100 Mm3. pumped from the river to a tank prior to being The Central Arizona Project (CAP), the fed to the vineyard irrigation bore. Hydrogeology steward of Arizona’s Colorado River water conditions in many other parts of southern and entitlement of 2,700 Mm3/year, and operator of the eastern Australia are found to be suitable for 550 km long CAP Aqueduct, has water spreading managed aquifer recharge. recharge facilities with an aquifer storage capacity of 460 Mm3/year (Lluria 2009). These facilities consist of three projects near Phoenix, three USA near Tucson and one between the Colorado River water diversion point of the CAP Aqueduct Adoption of MAR has a long history in the USA. and Phoenix City. The latter facility, called the A very wet period suffered in Arizona in the Tonopah Recharge Project (TRP), has a capacity 1980s indicated the need to store surface water of 185 Mm3/year and is utilized predominantly surpluses by means of artificially recharging for the recharge of water credits for the states of drafted aquifers. The laws adopted in 1980 Nevada and California in accordance with a Tri- and 1986 established the legal framework for State agreement. According to this agreement, all MAR aspects, including ownership of the Arizona stores water allocations of the Colorado recharged water and were the base for important River for Nevada and California in the wet years 6 in exchange for its (Arizona’s) water allocations smaller water spreading recharge facilities in from the Colorado River in the drought years. In Arizona. dry periods, Arizona recovers its water allocation There are other examples from Mexico, Spain, from the Colorado River from the underground Nepal and other countries when groundwater storage at the TRP. The city of Tucson has two depletion was attempted to be resolved by MAR large water spreading recharge projects with a (Dillon 2005). The main lessons from the MAR total capacity of 185 Mm3/year (Lluria 2009). It experience in the above countries are: i) advance also has a 36 Mm3/year water spreading recharge planning of MAR can prevent negative impacts of operation called the ‘Sweetwater Recharge groundwater development; ii) there are a variety of Project’, which stores only reclaimed water MAR methods which can be selected depending underground. The Viddler Water Company, a on hydrogeological and socioeconomic conditions private corporation, operates a 123 Mm3/year of a target area; iii) simple methods of MAR have water spreading facility near Phoenix to bank to be a priority, although advanced methods water for sale in the future. The recharge units should also be considered; and iv) MAR inclusion are basins developed in abandoned agriculture into river basin water management can bring fields with a slow infiltration rate. There are many benefits both at the local and basin scale. MAR in the Fergana Valley Study Area (1) glacier-snow; (2) snow-glacier; (3) snow; and (4) snow-rain. The Naryn River, the Sokh River The Fergana Valley depression is the area spread and the Isfara River are of glacier-snow type. The between the mountains of Kuramin and Chatkal Karadarya River and its tributaries are of snow- on the north, Atoinak and Fergana on the east, glacier type. Over 55% of the irrigated soils are and Alai and Turkestan on the south (Lange prone to salinity, including 71,922 ha that is highly 1964). The Fergana Valley covers a central part saline. of the depression bounded by the outcrops of the Mirzaev (1974) specified three hydrogeological Mesozoic and Paleozoic formations. This study zones in the Fergana Valley: (1) groundwater is limited to the part of the Fergana Valley within natural recharge and transit (Zone A); (2) spring Uzbekistan with an area of 17,000 km2. The (Zone B); and (3) groundwater dispersion (Zone irrigated area totals to 897,000 ha. The climate C) (see Figure 1). is semi-arid with low quantity of precipitation Zone A represents the upper part of the and high summer temperatures. The annual fans of the small rivers in the Valley. The rivers precipitation rate varies from 100 to 200 mm in and canals supply groundwater which is deep the central part of the valley and increases to 300 in Zone A. Water-bearing deposits of Zone A mm in the piedmont areas. The mean average are represented by coarse shingle and gravel temperature is at 14 oC. The altitude increases deposits, forming favorable conditions for water from west to east from 330 meters above sea storage. These highly permeable deposits are level (masl) to 600 masl. The valley is filled gradually replaced by the loam and sandy loam with alluvial deposits of rivers washed out in the deposits on the periphery of the fan, which mountain zone. By source of supply, the rivers belongs to Zone C. Between these two zones of the Fergana Valley are divided into four types: there is a narrow Zone B, where groundwater 7 forms springs and discharges into the drain increases from Zone A to Zone C. Groundwater system (Figure 1). Transmissivity of the water- abstraction, which was at a maximum level bearing stratum increases from Zone C to Zone of 4.4 km3 in the beginning of the 1990s, had A, and varies from 50 to 16,000 m2/day. On the decreased to 2.7-2.8 km3 by 2005 (Mavlonov et other hand, groundwater level and soil salinity al. 2006). FIGURE 1. Hydrogeological zones in the Fergana Valley. Source: Mirzaev 1974. The Proposed MAR Strategy for the Fergana Valley The ‘excessive’ flow available in the Fergana from canal irrigation to groundwater irrigation Valley for MAR includes the following: in small rivers and upstream sub-catchments ● Winter flow of small rivers. The average flow and adoption of water-saving technologies of small rivers entering the Fergana Valley will intendedly preserve the in-stream flow from October 1 to April 1 is about 1,000 and thereby increase the groundwater winter Mm3/year (Ivanov Yuri, Head of Department, recharge. the Uzhydromet, Tashkent, Uzbekistan, and ● Hydropower releases from the upstream Consultant to IWMI, pers. comm. 2009). At reservoir on the Naryn River. The shift in present, this flow, which is partially used for the beginning of the 1990s of the upstream agriculture, supplies groundwater and forms reservoir operations from irrigation to a the return flow to the Syrdarya River. Shifting hydropower generation regime increased the 8 winter flow and reduced the summer flow of other aquifers. The alternative option is to adopt the upstream reservoir. There are no free water-saving technologies that will gradually create capacities in the downstream reservoirs for additional free subsurface storages in the aquifers storage of extra winter flow from the Fergana with favorable conditions for water banking. Saved Valley. Furthermore, the ice-cover of the river water can be used for improving water quality. flow in the downstream does not allow water This approach differs from the activities under to be delivered to the river delta and to the implementation in the Fergana Valley by different Aral Sea. This extra winter flow varies from development projects aimed to gain local benefits 2,000 Mm3 in low water to 3,000 Mm3 in high (Wandert 2009). The projects aim to lower water years (Mustafaev et al. 2006). the groundwater level by increasing drainage ● Precipitation in the natural recharge zone capacity. This way they increase the return flow of groundwater. The groundwater natural to the river in the winter season when there is recharge zone has an area of approximately shortage of free storages in the river downstream. 400,000 ha (Zone A on Figure 1), including The approach, proposed in this report, suggests irrigated and non-irrigated lands, where saving excessive winter flows in subsurface precipitation rate in winter is 126 mm, on horizons and recovering this water in summer. average. Total precipitation available for This approach can reduce evaporation, flow to groundwater recharge in Zone A is at 500 sinks and pollution – all for an overall regional Mm3/year. Since current groundwater recharge benefit. The proposed MAR implementation from precipitation in winter is estimated at strategy consists of several steps: 100 Mm3/year, the adoption of appropriate ● Step 1: i) Assessment of potential for MAR technologies of soil tillage, crop selection and in the Fergana Valley aimed to determine water harvesting may significantly increase the subsurface free water storage available, groundwater recharge. or enhanced storage created by intensive ● Subsur face f low f rom the ups t ream . groundwater abstraction; ii) determining Subsurface flow from the upstream irrigated appropriate technologies for MAR; and iii) land occurs along the valleys of small estimating irrigated areas that have the rivers and is estimated to be 950 Mm3/year potential to shift from canal irrigation to (Mavlonov et al. 2006). Since the groundwater conjunctive use and considering the adoption level is shallow in half of the study area of potential water-saving technologies. (Dukhovny et al. 2005), most part of the ● Step 2: MAR activities in one of the pilot summer subsurface flow discharges into the aquifers should spread along the main canals. drainage system and enters the Syrdarya Since the groundwater level is high in the River in winter. canal command areas, it is appropriate to The water resources available for MAR make start MAR implementation by intensifying the 13-17% of the total inflow to the Fergana Valley, groundwater abstraction for irrigation purposes amounting to 24,600-28,300 Mm3/year in low and lowering the groundwater level. Then and high flow years, respectively. MAR will allow focus on storing winter flow of the Naryn increasing groundwater abstraction from 2,700 River and small rivers in the subsurface to 5,000-5,500 Mm3/year, mainly for irrigation horizons. At the same time, create incentives purposes (Mavlonov et al. 2006). Implementation for farmers to adopt water-saving irrigation of this strategy at the regional scale may require technologies in the river upstream to reduce different technologies. Simple structures, such saline fluxes from topsoil to groundwater. The as infiltration basins and percolation from the shift from canal irrigation to conjunctive use in riverbeds and floodplains, can be used in some the Fergana Valley will increase the summer of the aquifers, while deep underground dams are flow of the Naryn River for downstream use. the only option for subsurface water banking in Under new conditions, power stations can be 9 installed in main canals to produce power for lowering the water level, recharge structures the operation of wells. such as infiltration basins, boreholes and shafts can be constructed along the canal for MAR activities should be initiated in the storing the winter flow of the Naryn River in Isfara River Basin located in the tail end of the subsurface horizons. Then the capacity of the Big Fergana Canal (BFC) as it is easy to the power stations on the main canals can be estimate the impact of MAR and make the increased to produce energy for the operation necessary refinements in the Isfara Basin. of wells. Then move to the next subbasin Then move to the next subbasin along the along the BFC. BFC, which is the Sokh River Basin. ● Step 4: When the objective is achieved for ● Step 3: Shift from the canal irrigation to all separate aquifers along the main canals, conjunctive use in the Sokh River Basin consider MAR at the regional scale within the and adopt water-saving irrigation (including whole Fergana Valley. improved fur row and advanced dr ip) technologies in the river upstream and the The following sections describe the first three midstream. In the upstream, low barriers, steps, progress achieved and results obtained to proposed across the riverbed will increase date. Step 4 is not yet considered here, as it has the groundwater recharge. This groundwater to include more advanced stages of technology recharge will contribute to maintaining water development and has to capitalize on the success quality and storage. In the BFC zone, after of steps 1-3. Assessing MAR Potential in the Fergana Valley Methods and Data (Zone A) and in the main canal commands with transmissivity of the water-bearing stratum above The MAR potential in the Fergana Valley is 300 m2/day and groundwater level below a 3 m eva lua ted cons ider ing fac to rs , such as depth. Sources available for MAR are (1) free potential for water storage, depth to water winter flow of small rivers; (2) the flow of small table, groundwater sal inity, availabi l i ty of rivers, which can be released by the adoption excess water and other factors. Areas selected of water-saving technologies or increasing with favorable conditions for water storage groundwater irrigation; (3) precipitation in Zone include parts of small r iver basins where A; (4) subsurface inflow from the upstream; and free subsurface capacities are available, and (5) the winter flow of the Naryn River. Winter areas where free capacities can be created by flow of small rivers can be used for increasing intensive groundwater abstraction or reducing natural recharge in Zone A, which spread above the groundwater recharge by the adoption of the main canal commands. Natural recharge can water-saving technologies. The last approach be enhanced by increasing the leakage from the contributes to decreasing groundwater salinity by riverbed and the floodplain, canal and stream reducing saline fluxes from the vadoze zone and channels. The winter flow of the Naryn River increases the river free flow in summer, which is can be stored underground by: a) increasing the available for groundwater recharge. The areas leakage from the canals; b) installing infiltration suitable for groundwater storage were defined basins; and c) boreholes or shafts. Open drains, in the groundwater (GW) natural recharge zone after lowering the water table, may be used 10 under favorable geology conditions as recharge Groundwater irrigation is proposed for the structures as well. area with the transmissivity (T) of deposits above Areas suitable for groundwater irrigation or 300 m2/day, the groundwater level is less than conjunctive use may be specified within each 3 m depths and the salinity less than 2,000 hydrogeological zone based on transmissivity of mg/l. Conjunctive use of groundwater and canal subsurface horizons, water depths and quality water is recommended for the area with T > 300 (at first approximation – salinity) in the following m2/day and salinity less than 4,000 mg/l. The rest order: of the area is kept under canal irrigation. This 1) Subdistrict or hydrogeological zone (See area has a groundwater level below 12 m and/ Figure 1). or T < 300 m2/day. Single wells are proposed for the area with 100 < T < 300 m2/day. Groundwater 2) Blocks, or part of a subsurface horizon, irrigation area was specified using the data of selected on the basis of the transmissivity of the Institute of Hydrogeology and Engineering the water-bearing stratum in the top 0-100 m Geology and the Institute UzGIP (Mavlonov et layer and categorized into several groups: al. 2006; Khasanhanova et al. 2006). Using ● blocks with poor transmissivity of deposits these data, several GIS themes were created, less than 100 m2/day; such as: hydrogeological zones, specific water yield, transmissivity of the deposits, depth of ● blocks with low transmissivity from 100 to the groundwater level, groundwater salinity, etc. 300 m2/day; Groundwater budgets were compiled for low ● blocks with good transmissivity from 300 (2001) and high (1995) water years for each to 1,000 m2/day; and aquifer of the Fergana Valley to determine the ● blocks with high transmissivity above potential of groundwater abstraction within the 1,000 m2/day. selected areas. 3) Subblocks selected on the basis of the depth of the groundwater level are as follows: Results ● Subblocks with the groundwater level less than 3 m in depth from the ground The data given in Table 1 indicates that free surface; capacities exceeding 3,000 Mm3 in Zone A are available for storing the winter flow of small ● Subblocks with the groundwater level rivers, which varies within a range of 1,000-1,200 ranging from 3 to 7 m in depth; Mm3/year and are predominantly allocated for ● Subblocks with the groundwater level winter crop irrigation. The indicated area is located ranging from 7 to 12 m in depth; and at higher altitudes above the commands of the main canals, which deliver water from the Naryn ● Subblocks with the groundwater level River to water-short areas of the Fergana Valley. deeper than 12 m in depth from the Free capacities available and those that potentially ground surface. can be created within the main canal commands 4) Micro-blocks separated on the basis of the are illustrated by Figure 2 and Table 1. salinity of groundwater: The data given in Table 2 indicate free ● Micro-blocks with salinity less than 2,000 subsurface capacities in the zone of the main mg/l; canals, available for water banking, totaling 760 Mm3. Additional capacities which can be released ● Micro-blocks with salinity ranging from by lowering the groundwater level are estimated at 2,000 to 4,000 mg/l; and 186 Mm3 per meter of groundwater level drawdown. ● Micro-blocks with salinity above 4,000 These data show availability of subsurface horizons mg/l. for storing the winter flow. However, detailed 11 TABLE 1. Free capacities of the subsurface horizons of the Fergana Valley. Aquifer Recharge zone Areaа Free capacity (ha) (Мm3) Almaz-Varzyk 19,825 231 Kukumbai 2,658 54 Kasansai 4,351 30 Iskovat-Pishkaran 19,439 359 Sokh 34,589 1,452 Altyaryk-Beshalysh 7,366 28 Namangan 5,196 77 Isfara 4,385 90 Mailisu 17,513 22 Karaungur 3,944 5 Naryn 28,393 167 Chust-Pap 7,936 147 Andijan-Shahrihan 7,919 16 Chimien-Aval 3,651 88 Osh-Aravan 21,223 324 Nanai 4,349 71 Total 192,737 3,161 Source: Mavlonov et al. 2006. Note: аThe area within the recharge zone where free capacities are available. modeling and economic analysis are required Total groundwater recharge in Zones A and B to estimate the optimal level of groundwater (Figure 1) is estimated to be in the range of abstraction and recharge. MAR has to be preceded 5,624-6,005 Mm3/year in low and high water by increasing the groundwater abstraction to lower years, respectively. the water table. The areas suitable for groundwater Expanding the area under conjunctive use and irrigation and conjunctive use in the Fergana Valley the adoption of water-saving technologies (Figure are illustrated in Figure 3a. 3b) will decrease the groundwater recharge in The estimates show that the area suitable summer due to reducing losses from canals and for groundwater irrigation totals to 290,000 ha irrigated fields. Recharge deficit (~1,000 Mm3/year) and 243,000 ha for conjunctive use. The rest of can be compensated using the winter flow of the the area can be kept irrigated using canal water. Naryn River and small rivers. The data given above The potential volumes of groundwater extraction indicate the potential for MAR at the regional level depend on hydrogeology conditions (Zones A and and the next step is assessing the MAR potential B) and the replenishable groundwater resources. at the pilot aquifer level. 12 FIGURE 2. The areas with favorable hydrogeology conditions for storing winter flow of the Naryn River. Source: Karimov et al. 2010. Notes: BAC – Big Andijan Canal; BFC – Big Fergana Canal; NFC – Northern Fergana Canal; BNC – Big Namangan Canal. TABLE 2. Available and potential capacities within the Fergana Valley for storing winter flow of the Naryn River. Aquifers Source of GW Areaа Free Potential capacity (per recharge (canal) capacities meter of groundwater level drawdown) hа Mm3 Mm3/m Naryn BFC 36,859 158 37 Naryn BAC 24,440 52 24 Naryn NFC 23,769 85 24 Naryn NFC 20,228 181 19 Namangan BNC 5,371 77 5 Mailisu BFC 20,547 5 20 Andijan-Shahrihan BFC 4,443 0 4 Altyaryk-Beshalysh BFC 17,171 0 17 Sokh BFC 27,561 126 28 Isfara BFC 8,828 85 8 Total 189,217 769 186 Notes: аThe area where free capacities can be created by lowering the groundwater level; BAC – Big Andijan Canal; BFC – Big Fergana Canal; NFC – Northern Fergana Canal; BNC – Big Namangan Canal. 13 FIGURE 3. The area with favorable conditions for: a) groundwater irrigation, and b) adoption of water-saving technologies. а) b) Pilot MAR: The Isfara River Basin Site Description recharge of the Isfara Aquifer. Gravel and shingle deposits (more than 100 m thick), representing The Isfara River originates by the melting of the upper part of the river basin, form favorable glaciers and snow on the Alay Mountains. Long- conditions for groundwater recharge. To the north term average discharge of the river is at 14.7 of the BFC, the gravel and shingle deposits are m3/s. The BFC crosses the aquifer in the upper gradually replaced by loams and sand loams. part of the basin, which allows the use of simple The groundwater has a subsurface outflow to structures for recharging the surface water into the northeast of the Syrdarya River and to the the subsurface horizons. Leakages from the northwest of the Kairakum Reservoir. Salinity BFC, the riverbed and streams and widespread of the groundwater is less than 1,000 mg/l in canal system are the main sources of the the upper part, and 1,000-3,000 mg/l on the 14 periphery of the basin with some isolated spots nearest weather station. The concentration of the where groundwater salinity exceeds 3,000 mg/l. suspended sediments was analyzed in laboratory Irrigated soils spread in the upper part of the conditions using the de-silting method. Soil basin to the south of the BFC (that receive samples were collected from the bottom of the water from the Isfara River, and water lifted trench before and after the groundwater recharge from the BFC in part) and soils located on trial from depths of 0-25, 25-50, 50-75 and 75-100 the periphery of the basin are irrigated from cm below the ground. Soil samples were taken to the BFC. Groundwater extraction for irrigation determine particle size distribution, total dissolved purposes is 53 Mm3/year against a much higher solids, gypsum and carbonates by using standard 600 Mm3/year from the canal system. In the methods applied in the region (Arinushkina 1970). upper part of the basin, farmers grow mostly The second stage of MAR was carried out orchards and intercrops, such as vegetables, from March 26 to April 26, 2011. Unlike in the legumes, maize and sorghum for silage; whereas first stage, two de-silting basins were built of in the downstream in the canal command, they 3 m diameter and 1 m water depth before the grow mostly cotton and winter wheat. infiltration basin. Three monitoring wells, equipped with divers to get hourly observation data, were also installed 3, 30 and 35 m, from the north Field Experiments of Artificial Recharge of the basin. The divers were installed on April 5 and dismantled on May 31, 2011. Before the Methods and Data second stage experiment, sediments accumulated at the bottom of the basin were collected and A simple method of groundwater recharge was transported to the closest fields. Infiltration rates applied in the Isfara River upstream. One of were measured in three replications before widespread depressions along the BFC, a trench and after the experiment using the method of of 40 m x 25 m x 2 m size, was used in this infiltration rings. The infiltration rate was measured study as an infiltration basin (Figure 4). The soil in the beginning, middle and in the end of the profile of the selected site was presented by infiltration basin at 1, 5, 10, 15, 30, 45 and 60 shingle and gravel deposits filled with sand, and minutes after starting the experiment, and then was representative for the area along the BFC after each 30 minutes. The measurements in the river upstream (Figure 4c). The recharge continued until stable infiltration rates were study was carried out in two stages. In the first achieved, which were from 6.5 to 8.5 hours. stage, from April 1 to 17, 2010, water of the BFC After testing the MAR at the pilot field scale, was infiltrated underground from the infiltration groundwater modeling was applied to estimate basin of 0.1 ha area. Before the trial, the walls the water banking potential of the entire Isfara and the bottom of the trench were leveled. Two River Basin. water level meters were installed at the bottom of the trench. The water was entering the basin without initial treatment. Results and Discussions The discharge of the water entering the trench was measured using a fixed channel. Stage 1. Starting from April 1, 2010, water of Measurements were done every hour in daytime the BFC was supplied to the infiltration basin. and three times during nighttime. Altogether, 11 Due to the high percolation rate, the bottom of monitoring wells, located 100-1,000 m away from the basin was covered by water only from day 3 the infiltration basin, were installed to monitor the after starting the experiment. The water level in change of the groundwater level. Evaporation the basin stabilized on day 5 at 58 cm and was from groundwater was measured using the continuous for the next 10 days. After interrupting pan evaporator installed next to the infiltration the water supply, the water disappeared in the basin. Precipitation data was obtained from the basin after 3 days. Figure 5 shows that the 15 FIGURE 4. Artificial recharge of the Isfara Aquifer using water of the BFC (April, 2010): a) scheme of the study area; b) infiltration basin; and c) soil profile. (a) (b) (c) 16 infiltration rate had maximum values in the first In total, 40,000 m3 of water was supplied 3 days, when it exceeded 4.5 m/d, and then was into the basin, of which 2,000 m3 evaporated stable at 2-3.5 m/d starting from day 4 up to the and 38,000 m3 infiltrated to the groundwater. end of the experiment (Figure 5). Small regular This was a significant amount of water infiltrated variations of the water level can be explained from a small basin. Since the length of the canal by the high speed of the water entering the within the Isfara River Basin is 15,000 m, about basin, especially in the first 3 days during the 150 similar infiltration basins can be constructed initial stage of infiltration, forming waves and along the BFC for groundwater recharge. The due to high turbidity of the supplied water. Daily potential of full-scale MAR using these structures observations of the water levels showed that the is modeled. groundwater level rose by 35-45 cm at a distance To avoid over-estimation of groundwater of 250 m from the infiltration basin (Figure 6). recharge potential at the upscaling stage from FIGURE 5. Infiltration rate (m/d) during the recharge experiment in the Isfara River upstream. 7.5 y = 0.0008x4 - 130.14x3 + 8E+06x2 + 2E+11x + 2E+15 R² = 0.734 6 4.5 3 1.5 0 3/28/2010 4/2/2010 4/7/2010 4/12/2010 4/17/2010 FIGURE 6. Changes in groundwater level due to recharge from the infiltration basin. 17.55 17.7 17.85 18 18.15 18.3 18.45 4/3/2010 4/9/2010 4/15/2010 4/21/2010 4/27/2010 5/3/2010 5/9/2010 Note: Well 1*- located 250 m far from the infiltration basin. 17 Water depth (m) Infiltration rate (m/d) single point (0.1 ha) to the sub-catchment (15 ha), Soil texture given in Table 4 for November the recharge rates used in the modeling were 1.5 2010 indicates the content of the silt and clay times to twice as less when compared to values particles in the soil profile after the recharge trial found at the pilot site. carried out before the rainy season. As seen from The pilot GW recharge study indicated the the Table, the deposits form a separate layer on risk associated with high turbidity of the BFC the topsoil and can be easily removed from the water (Table 3). Data given in Table 3 shows that basin after completing the MAR trial. The data the sum of silt particles (diameter from 0.002 to given for March 2011 indicates soil texture after 0.05 mm) and clay particles (diameter less than the rainy season. A comparison of the content of 0.002 mm) exceeds 95% of the total suspended the silt and clay particles in the soil profile in the particles contained in the water of the canal. This infiltration basin before and after the rainy season data indicates a possible risk of soil porosity being indicates an increase in the content of the silt and blocked. However, the analysis of the particle size clay particles in the soil layer by 0-25 cm in the distribution indicated that the major part of the silt winter season (Table 4). If during the groundwater and clay particles is deposited at the bottom of recharge trial, clay particles accumulated mainly the basin and above the topsoil, and much less in above the topsoil, the rainy season changed the the soil profile. The thickness of the deposits was distribution of the fine particles in the soil profile. 10 cm in the head part, 7 cm in the middle and 3 In spite of the high content of the fine particles in cm in the tail end of the basin (Table 4). the water of the BFC, their movement in the soil TABLE 3. Mechanical composition of the soil samples from the bottom of the infiltration basin. Sampling point Depth November 2010 March 2011 (cm ) Particles (%) Sand Silt Clay Clay Suspended particles, BFC 5.6 69.4 25.0 Controla 0-10 57.2 34.9 7.9 5.6 10-25 77.6 18.2 4.5 2.6 25-35 84.8 12.3 2.9 0.7 35-50 85.0 12.3 2.9 0.7 Infiltration basin Deposits 11.7 76.3 12 15.3 Head part 0-10 85.2 11.9 2.9 3.2 M = 10 cm 10-25 88.4 9.2 2.4 9.6 25-35 89.7 7.9 2.4 1.5 35-50 86.8 10.6 2.6 1.5 Middle part Deposits 11.1 67.5 21.4 33.0 M = 5 cm 0-10 85.0 12.0 3.0 7.9 10-25 86.8 10.6 2.6 7.6 25-35 87.2 10.1 2.7 0.6 35-50 92.2 6.9 0.9 0.6 Tail end Deposits 15.4 62.3 22.3 36.4 M = 3 cm 0-10 79.2 16.1 4.7 10.4 10-25 83.9 12.8 2.9 0.5 25-35 85.6 12.1 2.3 0.8 35-50 85.6 11.3 2.1 0.8 Notes: aStone particles were removed before analysis of the mechanical composition of the soil; M – thickness of the deposits in the infiltration basin. 18 TABLE 4. Mechanical composition of the deposits on the bottom of the infiltration basin (2011). Basin Particles (%) Sand Silt Clay 1-0.05 mm 0.05-0.002 mm < 0.002 mm Head part (M = 18 cm) 4 69 27 Middle part (M = 7 cm) 5 55 40 Tail end (M = 2 cm) 5 54 41 profile had taken place mainly during the rainy the content of the suspended particles was still season after the completion of the groundwater high in the head part of the basin. An analysis recharge experiment. This data emphasizes the of the mechanical composition of the suspended need for removing the deposits from the infiltration particles in the water of the BFC showed that 68% structure before the rainy season. of the particles was from silt and 31.3% from clay Stage 2. The second experiment on the particles. In the second stage of the experiment artificial recharge was carried out from March in 2011, the thicknesses of the deposits were 18 26 to April 26, 2011. The concentration of the cm in the head part and only 2 cm in the tail end suspended particles was 2,030, 1,887, 62 and 30 of the basin (Table 3). mg/l in the water of the BFC, in the second de- In total, from March 26 to April 26, 2011, silting pit, in the middle and in the tail end of the the volume of the water infiltrated from the infiltration basin, respectively. Towards the end basin to the groundwater was 20,200 m3, and of the experiment, the content of the suspended evaporation from the groundwater level during the particles was 807, 783, 327 and 43 mg/l in the experiment was estimated at 139 m3. The rise of BFC, in the second de-silting pit, in the head the groundwater level monitored at the monitoring part and in the tail end of the basin, respectively. well equipped with the diver is given in Figure 7. It was noted that in spite of the availability of Figure 7 shows that groundwater level is the de-silting pits before the infiltration basin, raised by 30-35 cm in the well located next FIGURE 7. Changes of groundwater level as affected by infiltration from the basin (March – April, 2011). 21.6 21.65 21.7 21.75 21.8 21.85 21.9 21.95 22 0 250 500 750 1,000 1,250 Hours Note: Well 2*- located 3 m far from the infiltration basin. 19 Water depth (m) to the infiltration basin. The groundwater level MODFLOW software (Waterloo Hydrogeologic rise was at 10 cm at the wells located 30 and Inc. 2000). Visual MODFLOW is a widely 35 m from the basin. Water budgeting studies used Microsoft Windows-based version of the confirmed that infiltration rate during stage 2 US Geological Survey 3-D Finite Difference was less than stage 1, despite the duration of Groundwater Flow Model, MODFLOW (Harbaugh the second stage being longer. The volume of and McDonald 1996). The Isfara Aquifer Model groundwater recharge in stage 1 was twice that covers approximately 380 km2. Grid spacing in the of stage 2. This was caused by: i) smaller flow x and y model dimension is 50 m x 100 m, and in discharges entering the basin, and hence less the areas with dense irrigation canals and drainage water heads in the basin; and ii) late removal of ditches the model has 50 m x 50 m resolution. the deposits from the infiltration basin – after the The model boundary conditions were set rainy season. The results of the study suggest based on the results of the hydrogeological the importance of removing the deposits after studies carried out by the HYDROENGEO completing the recharge trial without delay (Miryusupov, Chief Hydrogeologist, Institute of before the rainy season sets in. In spite of Hydrogeology and Engineering Geology, pers. these limitations it was found that the recharge comm. 2010). The surface of the groundwater structures that have been tested have good level acted as a recharge boundary. The loamy/ potential to be used for water storing along the clay layer that is 300 m deep was set as a no- BFC. This concept was further tested through the flow boundary to represent the lower boundary groundwater modeling. condition. In the south, there is the subsurface inflow from the uplands through the valley of Modeling MAR the river. The groundwater level in the northeast is sourced by the Syrdarya River and in the Model Description northwest it is provided by a constant head. There is a zone of natural groundwater recharge on the A three-dimensional model of the Isfara Aquifer south and a discharge zone to the north of the (Figure 8a) was constructed using Visual BFC (Figure 8b). FIGURE 8. Three dimensional model of the Isfara Aquifer: a) three dimensional view, and b) plan. a) b) Source: Karimov et al. 2012. 20 The model has eight layers; first, third, fifth data was collected for each 1-10 minutes at the and seventh layer are represented by gravel and beginning of the pumping and three times per day shingle deposits in the recharge zone and by at the end stages and at the remote well. The loam and sandy loam deposits in the discharge hydrogeology parameters were estimated using zone. Groundwater is unconfined in the recharge groundwater level drawdown and restoration data zone and confined in the discharge zone in through analytical solutions of the Theis equation. layers two to eight. Main canals in the upper part According to these estimates, transmissivity of are given in the model as a ‘recharge boundary the water-bearing deposits varies in the range of condition’ because of their deep groundwater 40-555 m2/day and specific yield from 0.13-0.22 level. Canals that spread in the discharge zone m3/m3 in the unconfined zone and at 0.0001 are given as a ‘river boundary condition’ because m3/m3 in the confined zone. they supply the groundwater in summer and drain it in winter. Recharge of a ‘boundary condition’ Model Calibration and Verification also includes percolation losses of precipitation and infiltration losses of irrigation water. The Simplified models using Visual MODFLOW were infiltration losses pattern depends on the soil compiled for each of the 13 wells exploited for type, crop and groundwater table that is given in pumping tests. The size of each model was the model on a monthly basis. The BFC in the 1,000 m x 1,000 m. The simplified models study area is 2 m deep and 5 m wide. The water were represented by eight layers, repeating the depth in the canal is 1.5 m and thickness of the layers of the main model of the Isfara Aquifer. deposits at the bottom is 0.3 m. Initial depths The model grid was non-uniform – 5 m near of the groundwater level were taken from the the well and was increased to 20 m closer to database of the Institute of the Hydrogeology and the border of the model. In total, the model Engineering Geology. The initial groundwater level had 100 rows and 100 columns. The boundary was 20 m deep in the recharge zone and 1 to 2 of the model was taken as the constant head m below the ground to the north of the BFC. considering that short-term pumping will not In i t ia l values of the parameters were affect the water levels at 500 m distance from determined from pumping tests, carried out the well. A low permeable clay layer that is 300 by the HYDROENGEO in the study area from m deep was taken as an impermeable layer 1980-1985. During that time 13 pumping tests to represent the lower boundary of the model. were carried out including 9 in the unconfined In the beginning, the models were run using zone and 4 in the confined zone. Location of values of the parameters, coefficient of filtration the monitoring wells was dependent on the and specific yield, determined from an analytical hydrogeological profile. For a uniform profile, the solution of the Theis equation. Later, the model number of the observation wells taken was 2-3 parameters were specified using WINPEST, in the upstream, 3-4 in spring zone and 4-10 included into the Visual MODFLOW package on the periphery of the basin with a confined (Waterloo Hydrogeologic Inc. 2000). Running aquifer. The pumping tests were carried out WINPEST was aimed to correct values of the with fixed yields of the wells so as to simplify parameters by increasing the convergence with the analysis of the obtained data. The yields actual data obtained during the pumping tests. were from 25 to 100 l/s and the groundwater The comparison of the actual and the model level drawdown by 3-4 m in the exploited well. calculated values of the water elevations showed The yields of the wells were selected to achieve a coefficient of correlation at 0.85-0.95. Based on quasi-stationary regime and groundwater level the values of the parameters obtained from the drawdown by 20 cm in the remote well after 5-10 WINPEST, the values of the coefficient of filtration days. Duration of the pumping test was 10-15 and specific yields were corrected. Subsequently, days in the unconfined zone and 15-20 days in the historical groundwater budget data, obtained the confined zone. Groundwater level drawdown by the HYDROENGEO Institute from April 1, 1981 21 to April 1, 1983, were used for model calibration. groundwater storages by storing the winter flow The actual values of the groundwater budgets and of 100 Mm3/year each 2 years of 3 starting from elevations were compared with the model simulation year 5 of intensive groundwater extractions. results. The comparison showed a high convergence. Infiltration basins are modeled along the BFC. The value of the coefficient of correlation was Location of the wells is given in Figure 9b. at 0.989. Changes in the groundwater budget (groundwater extraction, recharge and evaporation) Simulations were done for each scenario since 1980 were considered in the formulation of the for 13 years starting from 2011, and the water modeling scenarios. extraction regime was fixed under scenario 1, seasonal variations are considered under scenario Modeling Scenarios 2 and long-term variation under scenarios 3 and 4 (Figure 10). Four alternative water management scenarios were considered: Modeling Results ● Scenario 1 (Sc1). The baseline scenario Results of the modeling is shown in Figure 11 simulates actual trends in groundwater and indicate high groundwater levels under the extraction for irrigation. The groundwater current baseline scenario (Sc1) and forming the resources are preserved for domestic and free capacities under scenario 2. A significant industrial requirements as well as to cover lowering of the groundwater level under scenario irrigation water shortages. The groundwater 3 is the consequence of the intensive groundwater extractions are at minimum levels of 1.7 m3/s, extractions exceeding the groundwater recharge. while the number of wells is 190 (Figure 9a). The regime of f i l l ing and draw off of the subsurface reservoir is shown in Figure 12. ● Scenario 2 (Sc2). Conjunct ive use of Under scenario 1 (Sc1) of minimum extraction groundwater and canal water for irrigation. This levels of groundwater for irrigation, the subsurface scenario proposes groundwater development reservoirs are filled during summer and drawn off in for irrigation in the upper part of the system winter for subsurface outflow and discharge to the and irrigation from the BFC in the downstream. drain system. Intensive groundwater extraction for The wells extract the annual groundwater irrigation (Sc3) results in drawing off water levels in recharge in the summer season. The number summer and minor filling happening in the winter. of wells is 230, of which 40 are projected along This increases the risk of groundwater depletion and the BFC – 0.5-2 km from north and south. This degradation in quality due to saline fluxes from the scenario aims gaining local benefits – more Vadoze Zone and surrounding inter-fan depressions. water available for irrigation in the Isfara Basin, Managed aquifer recharge in scenario 4 sustains but with no water saving for the downstream of the groundwater storages and maintains the water the Syrdarya River. quality, since 100 Mm3 of freshwater will be stored ● Scenario 3 (Sc3). The groundwater extraction underground. Groundwater storages are depleted exceeds its annual recharge by 20% and is in summer by intensive groundwater extraction but aimed to lower the groundwater level on the replenished in winter by managed aquifer recharge. periphery of the basin and arrest the salinity This combination aims at sustaining groundwater and waterlogging issues. storages and quality in the long run (see Figures 11 and 12). ● Scenario 4 (Sc4). Managed aquifer recharge The water-saving effect of the alternative = Scenario 3 plus storing 100 Mm3/year of the strategies expressed in the reduction of the winter flow of the Naryn River in the subsurface nonproductive depletions is given in Table 5. aquifers. The stored water in winter is projected Data presented in Table 5 demonstrates the to be withdrawn for irrigation in summer. This dependence between the return fraction, a ratio scenario proposes long-term regulation of between water extraction to recharge, and free 22 FIGURE 9. Location of the wells in the Isfara River Basin under the: a) scenario 2, and b) scenarios 3 and 4. а) b) FIGURE 10. Groundwater extraction regime under different scenarios of groundwater management. 3 2.5 2 Sc1 1.5 Sc2 Sc3, Sc4 1 0.5 0 0 720 1,440 2,160 2,880 3,600 4,320 Days 23 Groundwater extraction (Mm3/day) FIGURE 11. Water elevations in the upstream of the Isfara River as affected by the different water management scenarios. 420 410 Sc1 400 Sc2 390 Sc3 Sc4 380 370 0 720 1,440 2,160 2,880 3,600 4,320 Days FIGURE 12. Regime of filling and withdrawing from the subsurface reservoir as affected by the alternative scenarios of the groundwater management. Days 2,880 3,240 3,600 3,960 4,320 4,680 100 50 0 Sc1 Sc3 -50 Sc4 -100 -150 24 Storage change (Mm3) Water table (masl) TABLE 5. Changes of the groundwater storages in the Isfara Aquifer as affected by different scenarios of the groundwater management. Items Sc1 Sc2 Sc3 Sc4 Mm3/year Areal recharge 182 182 191 291 Leakage from the BFC 33 39 50 38 Groundwater extraction 53 228 295 309 Return flow 130 37 37 58 Evapotranspiration from groundwater 60 41 20 35 Nonproductive depletionsa 76 30 33 51 Including evaporation 18 12 6 11 Return winter flow 58 17 27 41 Return fraction 0.25 1.03 1.22 0.94 Storage change (as compared to scenario 1) -58 -83 -45 Free capacities 90 148 173 135 Recovery efficiency 0.79 Notes: aNonproductive depletions considered are the part of the evapotranspiration from groundwater table for physical evaporation and flow to sinks. The flow to sink in this case is the return flow to the river in winter, when the downstream reservoirs are full and there are no free storages. capacities. Increasing the return fraction from 0.22 increase in the salt affected and waterlogged to 1.22 increases free capacities from 90 to 173 areas. In addition, this strategy will produce Mm3 (see Tables 1 and 5). high non-process water depletions including The resources stored in the subsurface evaporation, flow to sinks and pollution. The horizons under scenario 4 were used in the second strategy of seasonal regulation of the following way: 14% was used for irrigation in groundwater storage will result in the reduction summer; 15% contributed for transpiration from of non-process depletions for evaporation, flow shallow groundwater level; 21% contributed to the to sinks and pollution. However, regional benefits return flow to the river in the summer season; and of this strategy will be insufficient. Unregulated 38% was still available in the subsurface horizon. groundwater extraction (Sc3) may result in the Nonproductive depletions constituted 5% for degradation of water quality and drawdown of the evaporation and 14% for return flow to the river in groundwater level, especially in the upper part winter. Furthermore, decreasing the groundwater of the basin where the groundwater is of high storages under scenario 4 indicates the potential quality. Finally, MAR strategy will facilitate the for additional recharge. Recovery efficiency of the prevention of groundwater depletions by storing aquifer recharge was estimated at 0.79. up to 100 Mm3/year of the winter flow from the Modeling results indicated differences in Naryn River. This strategy aims to effect regional the realization of the alternative strategies benefits by reducing the winter return flow by 17 of the groundwater management. The first Mm3/year and storing 100 Mm3/year of the winter strategy of preserving the underdeveloped flow of the Naryn River in the subsurface horizons groundwater will result in expanding the area of the Isfara Aquifer. with high groundwater levels in the Isfara River Wide-scale adoption of the alternative downstream. Thus, it will also cause a further strategies requires different approaches. Farmers 25 growing cotton and wheat under the State uneconomical. Since groundwater and surface patronage, such as subsidized resources, water use is regulated under the same water including water, have little impetus to save law in Central Asia, there is a strong procedure irrigation water. In contrast, farmers growing to be followed to get special permission to market crops, such as orchards, grapevines and access groundwater. The special permission vegetables, in the upper part of the system are restricts the amount of water that can be pumped more inclined to get access to the groundwater. and, as such, creates the basis for preventing However, construction of wells may require a groundwater depletions. significant amount of their income. The use The third strategy focuses on long-term of low quality pumps, available in the local regulat ion of the groundwater storage by market, would be a high risk exercise and accumulating the excessive flow of the rivers in result in considerable losses for farmers who the subsurface horizons in winter and its recovery intend shifting to groundwater irrigation for in summer for irrigation. The modeling results growing perennial crops, especially during their indicate the regional benefits of this strategy. establishment stage. Therefore, under the first The shift from canal irrigation to conjunctive use strategy, there is a high risk of incurring losses will release the summer flow of the Naryn River for small farms, which are located in the water for downstream use. Low groundwater levels deficit zone and are attempting to get access to in summer due to groundwater extraction will the groundwater. reduce the area prone to salinity and waterlogging The second strategy can be adopted by and also the area suitable for high-value crops. allocating preferential credit to the farmers for Increasing winter recharge using the freshwater installing wells and to cover the operational of the rivers will contribute to sustaining the water expenses during the first year of the establishment quality and reducing the return winter flow from of the orchards and grapevines, when farmers do the study area. not have free resources to invest into groundwater Increasing groundwater irr igation may development. The benefit of this strategy is increase consumption of electricity in the Fergana lowering the risk of losses for small farmers Valley. However, there is a widespread area shifting to groundwater irrigation and increasing under lift irrigation, which consumes even more the area used for high-value crops. Intensive energy. Shifting to conjunctive use will reduce groundwater extractions in the upper zone may, the area under lift irrigation, and decrease with time, cause negative processes such as: consumption of electricity in those areas. In degradation of the water quality in the upstream addition, small power stations proposed to be and the surrounding area due to saline fluxes installed in the canal system can generate power from the vadoze zone; and groundwater level for the operation of wells in the summer and for drawdown which will make water extractions rural population needs in the winter. Pilot MAR: The Sokh River Basin Site Description southern part of the basin is represented by the belt of the elevations elongated in the latitudinal The Sokh River Basin extends over 183,738 ha direction with altitudes of 800-950 masl. The from the northern foothills of the Turkestan-Alay Sokh River crosses the elevations from south to mountain system till the Syrdarya River. The north by narrow deep valley. Then to the north 26 from the hills there is the fan formed by the river The depth to access groundwater varies from covering the main part of the study area. In the 72 to 116 m in the head of the system, and can northern part of the study area, the periphery be as little as 0.5 to 2.5 m below ground level in of the fan merges with the alluvial valley of the the discharge zone. More than 800 wells have Syrdarya River with altitudes at 354-362 masl been in operation since the 1970s, but have (Geintsc 1967). generally only been used to supply peak irrigation The Sokh River is fed by meltwater from water demand in the summer in the 1980s to glaciers, with a maximum flow in the summer 1990s. The aquifer in the lower part of the basin and a minimum in February, when the baseflow is locally confined or semi-confined due to the is almost 100% sourced from groundwater. The discontinuous layers of clay and loam (Figure 13b) head reach of the river across the alluvial fan (Miryusupov and Gracheva 2006). is a natural recharge zone, contributing 44.5% of mean annual flow to groundwater. The Sokh River supplies water for irrigation in the upper Field Studies of Natural Recharge part of the basin, where soil cover is represented by gravel and sand, and is also the main source Methods and Data of the groundwater recharge. The river flow is distributed into irrigation canals at the headwork The field study in the Sokh River Basin focused called Sarykurgan, built on the river right after the on estimating the leakage from riverbed and the elevations. ways in which it could increase. The gravel field The Sokh Aquifer, underlying the Sokh of 600 ha area in the upper part of the fan creates River Basin (Figure 13a), consists primarily a favorable structure for replenishment of the of unconsolidated shingle and gravel outwash groundwater (Figure 14). deposits. In the lower reaches of the river, The field study consisted of two parts: (1) a spring zone appears in the form of a 3-5 water budgeting studies carried out from June km wide spring line that runs parallel to and to October 2010 (Figure 14); and (2) long- slightly upslope of the BFC (Figure 13a, b). term data analysis of the river flow and water Groundwater naturally discharges directly into quality. Water budgeting studies included the the drainage system over a 5 km wide belt that measurements of: the river flow discharge at lies downstream (just to the north) of the BFC the Sarykurgan Headwork; the water intake to alignment. The flow paths to the discharge zone the left bank and the right bank canals; the river are almost vertical in the narrow spring zone flow at the downstream of the headwork; and due to an impermeable anticline that almost water diversions into the secondary canals. The intersects the surface (Figure 13b), and gives measurements were carried out three times per rise to surface ponding, which then evaporates day. The groundwater elevations were monitored or flows into the drains. once in 3 days by monitoring wells located along The water-bearing strata consist of upper the riverbed. Water samples were collected Quaternary (QIII), intermediate Quaternary (QII) once per month for chemical analysis, which and lower Quaternary (QI) deposits. These was carried out in the laboratory conditions deposits contain gravel and shingle with an inter- using standard methods applied in the region layer of loamy sand and loamy deposits. The (Arinushkina 1970). Annually, in May, the water gravel and shingle deposits predominate in the management organization builds a dam across southern part of the study area with increasing the river to increase the water heads and divert proportions of loamy sand and loamy soils in the the river flow to the right bank canal. The dam northern parts. The intermediate Quaternary (QII) is 600 m long and 3 m high. The water depth in layer is subdivided into (QII-1) and (QII-2) layers, the riverbed was measured three times per day with low hydraulic and high hydraulic conductivity, from June to September. Subsequently, a relation respectively. was found between the river flow at the headwork 27 FIGURE 13. The fan of the Sokh River. a) plan, b) longitudinal profile of the fan, and c) cross-directional profile of the fan. Source: Gracheva et al. 2009. FIGURE 14. The gravel field: a) in the upper part of the Sokh River Basin (May, 2010), and b) the scheme of the main irrigation canals. a) b) and the leakage from the gravel field on the river. Results and Discussion Then this relation was applied to estimate leakage from the river in the long run using the river flow A significant part of the river flow released to the discharge data at the Sarykurgan Headwork from downstream of the Sarykurgan Headwork supplies 1995 to 2010, collected from the archival data the groundwater (Figure 15). of the Syrdarya-Sokh Basin Irrigation System The field studies in 2010 found that the Administration. leakage from the riverbed in the downstream 28 averages 30-35% of river flow releases to the of dissolved solids in the groundwater (Table downstream (Figure 16). 6) . F igure 16 shows that re lat ive losses The leakage from the gravel field allows are inc reas ing a t sma l l d ischarges and for the maintenance of low concentrations stabilizing at 25-35% for discharges exceeding FIGURE 15. Flow of the Sokh River in the downstream of the Sarykurgan Headwork (Q) and leakage from the downstream gravel field (Qс). 180 150 Q 120 Q c 90 60 30 0 7/30 8/6 8/13 8/20 8/27 9/3 9/10 9/17 Dates Source: Karimov et al. 2012. FIGURE 16. Relation between the leakage from the river gravel field and the flow discharge in the downstream of the Sarykurgan Headwork. 90 75 y = 99.581x-0.297 R² = 0.7171 60 45 30 15 0 0 20 40 60 80 100 120 140 160 River flow (m3/s) Source: Karimov et al. 2012. 29 Losses (%) Flow (m3/s) TABLE 6. Changes in the salinity of the groundwater in the upstream of the Sokh River. Parameters Unit Head part Inter-fan depression 20.07.10 8.09.10 21.10.10 20.08.10 TDSа mg/l 417±68 226±55 281±33 790±215 HCO3 mg-equ/l 2.6±0.5 1.9±0.3 1.8±0.9 4±1.9 SO4 mg-equ/l 3.9±0.9 1.6±0.5 2.4±1.3 8.4±3.5 Cl mg-equ/l 0.6±0.2 0.5±0.1 0.5±0.1 0.6±0.5 Ca mg-equ/l 1.4±0.2 0.9±0.1 0.7±0.3 3.5±1.2 Mg mg-equ/l 1.9±0.3 1.5±0.3 1.7±0.5 4±1.2 Na mg-equ/l 3.7±0.9 1.7±1.1 2.3±0.9 5.5±.8 pH 8 8±0.2 7.8±0.1 8±0.1 THb mg-equ/l 3.3±0.5 2.4±0.3 2.4±0.8 7.6±1.3 CHc mg-equ/l 2.6±0.5 1.9±0.3 1.6±0.4 4±1.9 NCHd mg-equ/l 0.8±0.3 0.9±0.1 1.3±1.1 4.8±1.4 Notes: a Total dissolved solids; b Total hardness; c Carbonate hardness; d Non-carbonate hardness. 50 m3/s . The data presented in Table 6 given in Figure 17. Changes in the groundwater show a decrease in the concentration of the salinity in the study area from 1995 to 2010 dissolved salts in the groundwater during indicate tight relations between the river flow the high leakage from the riverbed from July and the groundwater. Data given in Figure 17 to September, and, thereafter, it begins to show that the losses from the riverbed in the increase again. These data represent trends summer varies from 98 Mm3 in low water years in the salinity change in the central part of the to 137 Mm3 in high water years. The salinity of river upstream. the groundwater, as and when affected by the The concentration of the dissolved ions is leakage from the riverbed, begins to decrease in much higher in the groundwater of the inter-fan the spring and continues to the fall (Figure 18). depressions. There are two main factors affecting The gradual increase in the share of the saline the quality of the groundwater of the aquifer: water in the groundwater budget indicates the i) the leakage from the riverbed contributes to need for measures to sustain the quality of the the sustenance of the water quality; and ii) the water. subsurface inflow from the inter-fan depressions There are at least two ways to sustain and the upstream and saline fluxes from the the water quality: i) to adopt water-saving topsoil, increase the concentrations of the technologies to reduce losses from the irrigated dissolved solids. fields and to increase the natural recharge from During the field studies carried out in 2010, the riverbed and other recharge structures; and it was found that when the river flow exceeded ii) to restrict irrigation in the upstream of the the transporting capacity of the main canals, it river. This concept of adopting water-saving is released to the headwork downstream. Using technologies for conserving water for enhancing the relation obtained for 2010, the leakage from natural recharge of groundwater was further the riverbed was calculated for 1995-2010 and is tested through MAR modeling. 30 FIGURE 17. The Sokh River flow (Q) at the Sarykurgan headwork and the river flow released to the downstream (Qrel). 150 120 90 60 30 0 1995 1998 2001 2004 2007 Years Inflow (Q) Release (Q ) rel FIGURE 18. Leakage from the Sokh riverbed and the salinity of the groundwater (mg/l). 30 2,000 24 1,600 18 1,200 12 800 6 400 0 0 1995 1998 2001 2004 2007 Years Leakage TDS (mg/l) MAR Modeling present model output. The work was developed in four stages: (a) schematization of geology and Model Description hydrogeologic conditions; (b) structuring the model and data collection for calibration; (c) calibration The Sokh Aquifer Model was developed using of the model to historical data; and (d) scenario Visual MODFLOW (v. 4.2), with extensive use development and modeling. Paper maps were of GIS (ArcView 9.1) to: (1) prepare the input scanned and converted to ArcView polygons data and link it directly to the model, and (2) to (Rindahl 2004) and a digital elevation model of 31 Leakage (m³/s) Discharge (m³/s) TDS (mg/l) the land surface was created from 1:50,000 scale Layer 5. The base of this layer is set at maps using the Gauss-Kruger (1942) coordinate 50 m above mean sea level and is system. The stream network was transformed impermeable. The impermeable bed into raster format for direct incorporation into is very deep in the study area and MODFLOW. Data was mostly sourced from the there was no reason to consider such a HYDROENGEO database and used to create massive stratum. The bottom of layer 5 is the following thematic layers: location and details taken as constant to simplify the model of monitoring and pumping wells; groundwater and reflects the geological conditions contours; elevations of the top and bottom of each of the study area. It was also assumed geological layer (in spreadsheets); and hydraulic that a boundary condition at 300 m conductivity of each geological layer. Wells deeper below surface will not significantly affect than 100 m are typically used for domestic and subsurface water abstraction from depths industrial water supply, as are those with screens of 40-100 m from the soil surface. placed at lower depths (>70 m), where water Groundwater in layers three, four and five are quality is better and cost of pumping is less confined with a specific storage of 0.0001 1/m. significant than for irrigation or drainage. As a The model is bounded on the north by general first step, the boundary conditions, layers and head conditions, governing drainage outflow, and their interconnections were specified. The area the western and eastern boundaries are zero- represented by the model covers 54.75 × 50.25 flow boundaries lying at the edges of the aquifer. km in a grid of 335 rows and 365 columns with a The upstream condition is a fixed flow boundary, fixed cell size of 150 × 150 m. The aquifer system representing the underground inflow. The Sokh is represented by three distinct geologic units — River and the BFC flow in northern and western QIII, QII and QI (Figure 13). On the basis of the directions, respectively, and were included in the earlier hydrogeology surveys (Mirzaev 1974), model to provide local recharge and drainage of each unit was assigned a horizontal and a vertical the groundwater. The aquifer was divided into six hydraulic conductivity and thickness. zones as shown in Figure 19b. The discharge The three geologic strata are represented as zone is divided into the spring discharge (zone 3), five layers in the model, as illustrated in Figure upwelling (zones 4 and 5) and dispersion (zone 19a and described below: 6). Zone 3 is a belt that runs from 3 km to the south of the BFC to 5 km to its north. Layer 1. Soil surface to 20 m below ground The BFC was included in the model as a level. At the head of the valley, the layer river boundary condition. Average groundwater contains no water, but in the valley the discharge downstream of the canal alignment groundwater level is 0.5-3.0 m below the is 1.99 m3/s compared to 6.77 m3/s along the ground level. spring-lines upslope of the BFC. The natural Layer 2. From the bottom of layer 1 to the surface leakage along branches of Sokh River base of stratigraphic layer QIII in Figure was included as a linear recharge. The natural 13b, typically between 280 and 350 m recharge rates vary from 3,600 to 43,200 mm/year above mean sea level. at the stream channels in the upstream, but then Layer 3. The elevation of the base of this in the transit zone the intensity of the recharge layer corresponds to the stratigraphic at the stream channels falls to 1,080-18,000 mm/ boundary between geologic units QII1 and year. In other areas, recharge from irrigated lands QII2 and varies from 253 to 218 masl. predominates. The natural groundwater recharge from precipitation is estimated at 36 mm/year. The Layer 4. The base elevation of this layer is groundwater discharge in the upwelling zone is marked by the stratigraphic boundary of represented as the inflow to a 3 m deep surface geologic units QII2 and QI1 and varies drain with a constant flow depth of 1 m. A total from 218 to 125 masl. of 773 wells were in operation during the study 32 FIGURE 19. a) The structure of the Sokh Aquifer Model, and b) balance zones. a) b) Source: Gracheva et al. 2009. Notes: Zone 1: Groundwater recharge zone; Zone 2: Groundwater transit zone; Zone 3: Groundwater spring discharge zone; Zones 4-5: Groundwater upwelling with direct discharge to drainage; Zone 6: Groundwater dispersion zone. period, of which 667 were for irrigation, 57 for Model cal ibration was conducted by a drainage and 49 for domestic needs. stepwise adjustment of the hydraulic conductivity of each layer. The detailed groundwater budget data collected by the HYDROENGEO from Model Calibration January 1977 to December 1978 was used for In 1977-1978, the HYDROENGEO Institute model calibration. The modeled heads for each carried out detailed water budgeting studies month from January 1977 to December 1978 (Miryusupov and Gracheva 2006). The results of were compared to actual recorded values and these studies were used to calibrate the model. other comparators, which included groundwater The observed groundwater level elevations from discharge to the drainage system, and subsurface 44 observation wells were used to create a inflow and outflow to each zone. The aquifer groundwater contour layer in ArcView 9.1 and tests had determined hydraulic conductivity for were then interpolated to provide values for the discharge zone to be in the range of 4 to 12 MODFLOW. The water budget data included all m/d for the first layer, with 4 m/d used as the inflow components, such as subsurface inflow, initial value. The values of hydraulic conductivity recharge from the BFC, riverbed and streams, in each layer were increased step-by-step and and accessions from irrigation and rainfall. This values that resulted in the best fit between was balanced by discharge data on either side observed and modeled water levels were retained. of the BFC, upwelling, subsurface tail flows to Vertical hydraulic conductivities were treated in a the Syrdarya River, groundwater flows to surface similar manner, using the observed and modeled drains and direct evaporation. It was observed flows between the consecutive layers as the that subsurface inflow was almost twice the objective function. Initially, the ratio of horizontal subsurface outflow. to vertical conductivity was assumed to be 10:1 33 in the recharge zone and 100:1 in the discharge discharge to drainage and discharge of irrigation zone. The reliability of the model parameters was tailwater to drainage. Calculated drainage flow assessed through the comparison of: observed (groundwater discharge to drainage) does not and modeled groundwater level elevations; account tailwater losses, which are significant in drainage flows; and evaporation losses from Zones 5 and 6, presented by lowlands. This may groundwater and canal seepage values. be a reason for the higher difference between Actual and modeled groundwater levels the measured and calculated values of the were compared at successive time intervals drainage of groundwater. of 30, 210 and 720 days after the beginning The corrected values of horizontal and vertical of simulation. Sample plots of observed and conductivity are given in Table 8. modeled groundwater level elevation over time Inter-layers of low permeability were not are given for water levels in three different initially included in the model and their effect zones in Figure 20a, b and c, which show an is accounted for through the determination of acceptable level of overall correspondence, but adjusted values of vertical conductivity. There was indicates that further calibration improvements no change in the boundary conditions since 1978. would be obtained through automatic calibration. Changes in the groundwater budget (groundwater However, the authors preferred to retain the extractions, recharge and evaporation) since 1978 manual calibration based on adjusted hydraulic were considered n the formulation of the modeling conductivities, since this has a physical meaning scenarios. and is related to their knowledge of the aquifer and its behavior. Modeling Scenarios The comparison of calculated and actual discharge of groundwater into the drainage Five scenarios were developed to support alter- network is shown in Figure 21, indicating a good native strategies of groundwater recharge and fit between observed and modeled data, within development in the Sokh River Basin: 10% overall. However, the overall drainage Scenario 1 (Sc1) – the groundwater extraction volume predicted by the model was approximately at a minimum level of 3.8 m3/s. The surface flow 22% less than what was observed, as shown is the main source of irrigation water under this in Table 7, and this is believed to be due to scenario. additional unaccounted surface flows entering the Scenario 2 (Sc2) – on-farm furrow irrigation drainage network. The amplitude of the recharge is improved by int roducing water-saving is typically 35-50 m3/s but the fluctuation of the technologies, such as mulches, and alternate and drainage flow is much less, indicating the high short furrows. It is expected that these measures regulation capacity of the Sokh Aquifer. will reduce the groundwater recharge from the The peak recharge registered in July varied upstream irrigated land by 20%. from 81 to 89 m3/s, but the peak in drainage Scenario 3 (Sc3) – the introduction of rate begins 3 months later in October and runs advanced irrigation technologies in the river for 3 months at around 20 m3/s. Finally, the total upstream. It is expected that this measure will volume directly evaporated from groundwater produce a 40% reduction of the groundwater amounted to 93.0 Mm³/year compared to the recharge in the river upstream. present value, estimated to be 100.0 Mm³/ Scenar io 4 (Sc4) – the groundwater abstraction at a maximum level of 22.4 m3 year, approximately 7% lower. Calculated /s. values of leakage from BFC were 82 Mm³/year The effect of introducing advanced irrigation compared to 97.7 Mm³/year, or a difference technologies in the river upstream on groundwater of 16%. Given the uncertainty in the original recharge is accounted for. estimates, both these values were considered Scenario 5 (Sc5) – the groundwater extraction to be acceptable for proceeding to simulation. level is the same as in scenario 4 plus increasing Measured drainage flow includes groundwater groundwater recharge in winter from the Sokh 34 FIGURE 20. Comparison of observed and modeled groundwater levels in three different zones: a) Zone 4, b) Zone 5, and c) Zone 6. a) 378 377.5 377 376.5 376 375.5 375 Jan 77 Apr 77 Jul 77 Oct 77 Jan 78 Apr 78 Jul 78 Oct 78 Observation wells 546 (Calculated) Observation wells 546 (Observed) b) 397.5 397 396.5 396 395.5 395 394.5 Jan 77 Apr 77 Jul 77 Oct 77 Jan 78 Apr 78 Jul 78 Oct 78 Observation wells 533 (Calculated) Observation wells 533 (Observed) c) 380 379 378 377 376 375 374 Jan 77 Apr 77 Jul 77 Oct 77 Jan 78 Apr 78 Jul 78 Oct 78 Observation wells 337 (Calculated) Observation wells 337 (Observed) Source: Gracheva et al. 2009. 35 Head (masl) Head (masl) Head (masl) TABLE 7. Comparison of predicted and actual values of drainage flow for the Sokh River Basin. Zone Annual drainage of groundwater Calculated Measured Difference (Mm³) (Mm³) (%) Zone 3 179.27 213.37 16 Zone 4 66.7 62.63 -6 Zone 5 43.96 72.22 39 Zone 6 50.52 64.55 22 Total 340.45 412.77 18 Source: Gracheva et al. 2009. TABLE 8. Corrected values of hydraulic conductivity. Layer Recharge zone (Zones 1 and 2) Discharge zone (Zones 3, 4, 5 and 6) Horizontal Vertical Horizontal Vertical Initial Final Initial Final Initial Final Initial Final m/d m/d m/d m/d m/d m/d m/d m/d 1 4-15 10 0.08-12.5 0.05-1.6 2 115-120 90-100 0.08-11.5 0.05-1.6 4-15 10 0.08-12.5 0.05-1.6 3 40-65 35-50 0.047-6.5 0.43-0.7 7-10 8 0.04-6.5 0.43-0.7 4 40-65 35-50 0.047-6.5 0.43-0.7 7-10 8 0.04-6.5 0.43-0.7 5 15-25 12-20 0.047-6.5 0.43-0.7 2-3 6 0.04-6.5 0.43-0.7 Source: Gracheva et al. 2009. FIGURE 21. Example of comparison of predicted and actual values of drainage flow at drain ‘K’ and recharge of groundwater. 100 80 60 40 20 0 Jan 77 Apr 77 Jul 77 Oct 77 Jan 78 Apr 78 Jul 78 Oct 78 Predicted Actual Recharge Source: Gracheva et al. 2009. 36 Drainage, Recharge (m³/s) River f loodplain by 200 Mm3/year and by will reduce nonproductive evaporation by 58% and transporting water of the Naryn River through the winter return flow by 46% (Table 9). However, this BFC (Figure 13a). strategy may, with time, deplete groundwater storage, which would affect the quantity and the quality of the Results and Discussion groundwater. The regime of filling and draw off of the subsurface aquifer is given in Figure 23. The results of the water level modeling are Under scenario 1, groundwater storage shown in Figure 22. Figure 22 indicates that is filled in the summer and draw off occurs in following the first strategy without (scenario 1) and the winter for return flow to the river. Under with (scenarios 2 and 3) on-farm improvements scenario 4 there is a minor drawdown in the would cause high groundwater levels in both the summer and filling is done in the winter. Under upstream and downstream parts of the basin, and the scenario of MAR (Sc5), the storages are high return flow to the Syrdarya River in the winter. filled in the winter and drawdown occurs in the Development of groundwater for irrigation summer. Groundwater extractions at the minimum purposes under the second strategy (scenario 4) levels simulated in scenarios 1-3 cause the FIGURE 22. Changes of the groundwater elevations in the: a) upper, and b) lower parts of the Sokh River Basin as affected different scenarios of the groundwater management. а) 426 424 422 Sc1 420 Sc2 Sc3 418 Sc4 Sc5 416 414 412 0 360 720 1,080 1,440 1,800 Days b) 383.0 382.5 382.0 381.5 Sc1 Sc2 381.0 Sc3 380.5 Sc4 380.0 Sc5 379.5 379.0 0 360 720 1,080 1,440 1,800 Days 37 Water level (masl) Water level (masl) FIGURE 23. Calculated regime of filling and draw off of the groundwater storages in the Sokh Aquifer. Days 0 360 720 1,080 1,440 1,800 150 100 50 Sc1 0 Sc4 Sc5 -50 -100 -150 TABLE 9. Changes of the groundwater storages under alternative groundwater management strategies in the Sokh River Basin (Mm3). Sc1 Sc2 Sc3 Sc4 Sc5 Annual Recharge 1,113 953 939 991 1,192 Leakage 68 80 81 125 124 Extraction 117 117 117 698 698 Return flow 609 540 533 302 322 Evaporation 470 405 399 198 205 Storage change 92 71 69 -18 118 Winter Recharge 288 246 244 253 359 Leakage 23 31 32 51 50 Extraction 0 0 0 107 107 Return flow 393 356 354 214 231 Evaporation 48 41 41 36 37 Storage change -91 -78 -78 -39 -45 Summer Recharge 825 707 694 738 832 Leakage 46 26 49 74 73 Extraction 591 117 117 591 591 Return flow 92 185 179 89 92 Evaporation 422 365 359 162 168 Storage change 121 100 99 -21 45 38 Storage change (Mm3) highest return flow to the river, which is estimated lowering of the groundwater levels in the BFC at 52% of the groundwater recharge, and zone increased leakage from the canal from 81 evapotranspiration from the groundwater level, to 125 Mm3/year, or by 50%. It is important to most of which is non-process and at 39-40% of note that if under scenarios 1 to 3, groundwater the total recharge. Groundwater storages increase storages are filled in summer and draw off by 92 Mm3/year under this strategy, and thereby occurs in winter, then under scenario 4, where increase the area with shallow groundwater level groundwater development is unregulated, draw on the periphery of the fan, followed by salinity off occurs in summer and winter; and under and waterlogging issues. scenario 5 of MAR, the groundwater storages are The shift from canal irrigation to conjunctive intendedly filled in winter and draw off in summer. use, modeled under scenario 4, will reduce the Under the MAR scenario the return flow in winter return flow to the river from 52% to 27% of total was successfully reduced from 393 Mm3 to 231 groundwater recharge and evapotranspiration from Mm3 per season (Table 10). 52% to 18%. Gradual lowering of the groundwater The modeling results given in Table 10 show level, with time, may cause groundwater depletion low efficiency of the groundwater management and affect its quantity and quality. The last may under strategy 1, which increases to 0.63 under be caused by saline fluxes from the vadoze zone strategy 4, with managed aquifer recharge and from the inter-fan depressions and from the and conjunctive use of groundwater and canal upstream of the Sokh River. water for irrigation. Non-process depletions Under scenario 5 of MAR, the return flow reduced from 48% under strategy 1 to 25% to the river is estimated to be 24% to 25% under strategy 3. Similar potential for MAR is of the total groundwater recharge and the available in other aquifers of the Fergana Valley, evapotranspiration from 13% to 16%. Storing which demonstrates the importance of this 200 Mm3/year of the winter flow of the Sokh strategy to improve water management in the River in the subsurface horizons increased the Fergana Valley and in the Syrdarya River Basin, groundwater extraction levels to 27 m3/s. The on the whole. TABLE 10. Main induces of the alternative strategies of the groundwater management in the Sokh River Basin (Mm3/year). Induces Sc1 Sc2 Sc3 Sc4 Sc5 Groundwater recharge 1,113 953 939 991 1,192 Including MAR using winter flow 115 of the Sokh River Naryn River 27 Groundwater extractions 117 117 117 698 698 Non-process depletions: 534 477 474 273 292 Evaporation 141 122 120 59 62 Return winter flow 393 356 354 214 231 Storage change -15 -15 -68 105 Free capacities 1,452 1,467 1,467 1,535 1,347 Resource recovery 0.10 0.11 0.12 0.63 0.53 39 Conclusions It is observed that there is a growing demand and by introducing water-saving technologies; for food and energy, and also an increased and winter precipitation at 500 Mm3/year. The competition for water between upstream and resources available for MAR make 13% to 17% downstream users in the Syrdarya River Basin. of the total inflow to the Fergana Valley in low to Furthermore, the change in the upstream reservoir high flow years, respectively. operation from a conjunctive irrigation/hydropower The study followed the stepwise procedure of mode to an exclusively hydropower generation implementing MAR in the Fergana Valley. The first mode reduced the flow of the river downstream step is the regional assessment of the potential in the summer and increased it in the winter. for MAR and for shifting from canal irrigation The coincidence of peaks in winter hydropower to conjunctive surface water-groundwater use. releases and return flow from the irrigated land The second step is the application of MAR for in the Fergana Valley forms excessive river aquifers, located in the tail end of main canals. flows, which complicates the operation of the The next step is to move to the next aquifers downstream reservoirs. As a result, there is along the main canals. When the process is a water shortage in the range of 2,000-3,000 complete for all of the separate aquifers along the Mm3/year affecting downstream water users in main canals, MAR implementation for the entire the summer and excessive, often unutilized, flows Fergana Valley is considered. of the same magnitude in the winter. Projected The regional assessments in the Fergana reduction of the river flow by around 6-10% by Valley show that over 500,000 ha or 55% of the 2050 due to climate change, with increased currently irrigated land can be shifted from canal frequency of extreme, high and low flows may irrigation to conjunctive surface water-groundwater further complicate downstream basin water use, which will reduce the return flow to the river management, which is currently accomplished by 30%, or by 1,000 Mm3/year, and form free primarily by a cascade of reservoirs. This study storages of 500 Mm3 in the command areas of suggests that the current practice of sequential the main canals. Pilot-scale field and modeling in-channel reservoirs is not coping well with the studies of MAR for the Isfara Aquifer, located in needs of both upstream and the downstream the tail end of the BFC, found that groundwater water users. extractions in the summer exceeding the annual The study further suggests that MAR in the recharge by 20% will create capacities for storing upstream of Fergana Valley and elsewhere in 100 Mm3 of winter flow of Naryn River in the the Syrdarya Basin may help adapt to a new Isfara Aquifer. water management reality. Over 3,000 Mm3/year Pilot-scale field and modeling studies of MAR of subsurface free capacity is available in the for the Sokh Aquifer, located next along the BFC, upstream of the small river basins of the Fergana indicated that the seasonal extraction of 63% of Valley. These capacities can be used for storing annual groundwater recharge and introducing excessive flows of small rivers and thereby water-saving technologies in the Sokh River effectively reduce the return flow to the Syrdarya. upstream will free the river winter flow of 115 Additional free storage for MAR can be created in Mm3 for enhancing natural recharge from the the command areas of the main irrigation canals riverbed. The increased groundwater extraction by intensive groundwater extraction. The water will reduce the return flow to the Syrdarya River resources available in the Fergana Valley for in the winter by 162 Mm3 and an additional 100 MAR include: winter flow of the Naryn River from Mm3/year of the summer flow of the Naryn River 2,000 to 3,000 Mm3/season; winter flow of small can be released for the Syrdarya downstream. rivers at 1,000 Mm3/year, which could be free by Overall, groundwater development for irrigation increasing groundwater extractions for irrigation and MAR in the Fergana Valley may reduce 40 the winter flow of the Syrdarya River at the northern Tajikistan and lowlands of the Fergana Valley outlet by 1,500 Mm3/year and southern Kazakhstan. consequently increase the summer flow of the ● Potential for adoption of advanced MAR river to the same magnitude. technologies, such as subsurface artificial The results of the study suggest that simple dams and aquifer storage and recovery technologies, such as infiltration basins and technologies. enhanced natural recharge from riverbeds, can be used for MAR in the Fergana Valley. Small ● Management of groundwater quality by MAR. infiltration basins can be constructed along the ● Adoption of water-saving technologies, such main canals, delivering the water of the Naryn as drip irrigation, and MAR. River to the water-short areas of the Fergana The results of the study propose (especially Valley. Enhanced natural recharge from river for projects already under implementation in the floodplains is found to be effective for sustaining region) shifting the focus from reconstruction of groundwater storages and preserving the high dense drainage systems in the Fergana Valley to quality of groundwater in small river basins. To groundwater development for irrigation and MAR. ensure the science-based implementation of MAR The ‘Meliorative fund’, established for amelioration at the scale of the Syrdarya River Basin at large, of salt-affected and low productive land and the following need to be examined: functioning under the Government of Uzbekistan, ● Potential for introducing MAR in the Syrdarya can also be used as a financial instrument for River Basin, including the foothills of the MAR activities. References Abdullaev, U.; Khasankhanova, G.; Myagkov, S.; Khamzina, T.; Agalceva, N.; Fazilova, M.; Lysenko, P.; Merkushkin, A.; Abdullaev, Sh.; Voronov, A.; Manuyk, N.; Nechaeva, G. 2007. Water - critical resource for Uzbekistan’s future. Tashkent, Uzbekistan: United Nations Development Programme (UNDP). 24p. Agaltsceva, N.A.; Pak, A.V. 2007. Impact of climate change on river flow of Aral Sea basin. In: Bulletin: Climatic scenarios, impact of climate change. Tashkent: Research Institute of Hydrometeorology. P. 44-52. Akramov, A.A. 1991. Regulating water resources in groundwater aquifers. Tashkent: Science of Uzbekistan. 207p. (in Russian). Arinushkina, E.V. 1970. Handbook for chemical analysis of soil (in Russian). Moscow, Russia: Publishing House of the Moscow State University. Central Ground Water Board. 2005. Master plan for artificial recharge to ground water in India. 117p. India: Central Ground Water Board. Available at http://cgwb.gov.in/documents/MASTER%20PLAN%20Final-2002.pdf (accessed on February 17, 2013). Charlesworth, P.B.; Narayan, K.A.; Bristow, K.L.; Lowis, B.; Laidlow, G.; McGowan, R. 2002. The Burdekin delta – Australia’s oldest artificial recharge scheme. In: Management of aquifer recharge for sustainability, Dillon, P.J., ed., Proceedings of the 4th International Symposium on Artificial Recharge (ISAR4), Adelaide, September 22-26, 2002, Swets & Zeitlinger, Lisse, ISBN 90 5809 527 4, pp. 347-352. Dillon, P. 2005. Future management of aquifer recharge. Hydrogeol. J. 13: 313-316. 41 Dillon, P. 2011. Managed aquifer recharge - examples of success. National Research Flagships. Commonwealth Scientific and Industrial Research Organisation (CSIRO). Available at http://www.bebuffered.com/downloads/ Action%20Research%20Worshop%2024%20Feb%202011%20Dillon%20CSIRO_LQ.pdf (accessed on February 17, 2013). Dukhovny, V.A.; Umarov, P.D.; Yakubov, Kh.E.; Tuchin, A.I. 2005. Drainage in the Aral Sea Basin – Towards a strategy for sustainable irrigated agriculture. In: International Conference Report on Towards a Strategy for Sustainable Irrigated Agriculture with Feasible Investment in Drainage, Aral Sea Basin, Central Asia, March 10- 13, 2004, Tashkent, Uzbekistan. pp. 134–162. EPA (Environmental Protection Authority). 2009. Guidelines for Managed Aquifer Recharge (MAR) – Health and environmental risk management. EPA Victoria Publication 1290, July 2009. Available at http://epanote2.epa.vic. gov.au/EPA/publications.nsf/PubDocsLU/1290? OpenDocument (accessed on February 17, 2013). Geintsc, V.A. 1967. Underground waters of quaternary deposits of south-western Fergana, their regime and budget in connections with water related activities. Tashkent: Phan. 98p. (In Russian). GHD Pty Ltd. and AGT Pty Ltd. 2011. Feasibility of managed aquifer recharge for agriculture. Waterlines Report. Canberra, Australia: National Water Commission. Gracheva, I.; Karimov, A.; Turral, H.; Miryusupov, F. 2009. Assessment of the potential and impacts of winter water banking in the Sokh aquifer, Central Asia. Hydrogeol. J. 17(6): 1471-1482. Harbaugh, A.W.; McDonald, M.G. 1996. User’s Documentation for MODFLOW-96, an update to the US Geological Survey modular finite difference ground-water flow model. Open-File Report 96-485, USGS: Reston; 56 pp. Ishida, S.; Tsuchihara, T.; Yoshimoto, S.; Imaizumi, M. 2011. Sustainable use of groundwater with underground dams. JARQ 45(1): 51-61. http://www.jircas.affrc.go.jp Jia, Y.; You, J. 2010. Sustainable groundwater management in the North China Plain: Main issues, practices and foresights. In: Proceedings of XXXVIII IAH Congress. Groundwater Quality Sustainability, Krakow, 12-17 September 2010, Extended abstracts, eds., Zuber, A.; Kania,J.; Kmiecik, E.P. 855-862. Jinxia, W.; Huang, J.; Rozelle, S.; Huang, Q.; Blanke, A. 2007. Agriculture and groundwater development in northern China: Trends, institutional responses, and policy options. Water Policy 9(Supplement 1): 61-74. Karimov, A.; Mavlonov, A.; Miryusupov, F.; Gracheva, I.; Borisov, V.; Abdurahmonov, B. 2012. Modelling policy alternatives toward managed aquifer recharge in the Fergana Valley, Central Asia. Water International 37(4): 380-394. Karimov, A.; Smakhtin, V.; Mavlonov, A.; Gracheva, I. 2010. Water ‘banking’ in Fergana valley aquifers—A solution to water allocation in the Syrdarya river basin? Agricultural Water Management 97(10):1461–1468. Khasanhanova, G.; Khamzina, T.; Ibragimov, R. 2006. Crop water requirements in the Fergana Valley. Paper presented in the Regional Conference on: Conjunctive use of ground and surface water resources of Fergana valley for irrigation. November 2, 2006. Tashkent, Uzbekistan. Kumar, S.A.; Rajput, T.B.S. 2005. Optimizing water uses and recharging of aquifers. Indian farming. New Delhi: Indian Agricultural Research Institute. Lange, O.K. 1964. Hydrogeological and engineering geological conditions of Uzbekistan. Vol.II. Tashkent: Fan, 319 p. Lluria, M.R. 2009. Successful application of Managed Aquifer Recharge in the improvement of the water resources management of semi-arid regions: Examples from Arizona and the Southwestern U.S.A. Boletín Geológico y Minero 120(2): 111-120. Mavlonov, A.A.; Borisov, V.A.; Djumanov, J. Kh. 2006. Assessment of groundwater resources of Fergana Valley (in Russian). Paper presented in the Regional Conf. on: Conjunctive use of ground and surface water resources of Fergana valley for irrigation, November 2, 2006. Tashkent, Uzbekistan. Ministry of Land Resources of China. 2005. Report on groundwater resources and environment investigation in China. Beijing, China: Ministry of Land Resources. 42 Ministry of Water Resources of China and Nanjing Water Institute. 2004. Groundwater exploitation and utilization in the early 21st century. Beijing, China: China Water Resources and Hydropower Publishing House. Miryusupov, F.; Gracheva, I. 2006. Hydrogeology schematization of the Sokh aquifer (In Russian). Paper presented in the Regional Conf. on: Conjunctive use of ground and surface water resources of Fergana valley for irrigation, November 2, 2006. Tashkent, Uzbekistan. Mirzaev, S.Sh. 1972. Мethodology of regional evaluation of potential of groundwater artificial recharge. Uzbek Geol J. N1: 5-7. (in Russian). Mirzaev, S.Sh. 1974. Groundwater storages of Uzbekistan. Tashkent: Fan. Mustafaev, J.; Ryabtscev, A.D.; Balgerei, M.A.; Karlyhanov, O.K. 2006. Issues of releases of winter flow of Syrdarya down to Chardara reservoir. J. of Water Management of Kazakhstan N1: 2-8. Parsons, S.; Dillon, P.; Irvine, E.; Holland, G.; Kaufman, C. 2012. Progress in managed aquifer recharge in Australia. Waterlines Report Series No. 73, March 2012. Available at http://www.nwc.gov.au/__data/assets/ pdf_file/0015/21534/Waterlines-73-Managed-aquifer-recharge.pdf (accessed on February 18, 2013). Rindahl, B. 2004. Groundwater modeling and analysis using the USGS MODFLOW program and ArcView. Published Online in ESRI International User Conference 1996 Proceedings. ESRI International User Conference. http:// proceedings.esri.com/library/userconf/proc96/TO50/PAP042/P42.HTM (accessed on February 18, 2013). Romani, S. 2005. National blueprint for recharging groundwater resources of India. Central Groundwater Board. http://www.publications.iwmi.org/pdf/H039309.pdf (accessed on February 18, 2013). Rosegrant, M.W.; Cai, X.; Cline, S.A. 2002. World water and food to 2025: Dealing with scarcity. Washington, DC: International Food Policy Research Institute. Sakthivadivel, R. 2007. The groundwater recharge movement in India. In: The agricultural groundwater revolution: Opportunities and threats for development, Giordano, M.; Villholth, K.G. (eds). Wallingford, UK: CABI. 196-210. (Comprehensive Assessment of Water Management in Agriculture Series 3). Available at http://www.iwmi.cgiar. org/publications/CABI_Publications/CA_CABI_Series/Ground_Water Protected/Giordano_1845931726-Chapter10. pdf (accessed on February 20, 2012). Shah, T.; Molden, D.; Sakthivadivel, R.; Seckler, D. 2000. The global groundwater situation: Overview of opportunities and challenges. Colombo, Sri Lanka: International Water Management Institute (IWMI). Sharma, K.D. 2009. Groundwater management for food security. Current Science 96(11): 1444. Sherfetdinov, L.Z. 2000. Resources of groundwater of southern flank of Aral Sea basin. In: Water resources, Aral Sea issues and environment. Tashkent: University, P. 40-58. Tihonova, K.A. 1972. Using regulated subsurface reservoirs and pump stations for irrigation purposes for irrigation of big areas. Materials of the International Symposium on Water Management Systems. Prague. Tuinhof, A.; Olsthoorn, T.; Heederik, J.P.; de Vries, J. 2003. Management of aquifer recharge and subsurface storage - A promising option to cope with increasing storage needs. In: Management of aquifer recharge and subsurface storage: Making better use of our largest reservoir. Seminar Wageningen 18-19 December, 2002, eds., Tuinhoff, A.; Heederik, J.P. Netherlands National Committee (NNC) – International Association of Hydrogeologists (IAH) Publication No. 4. The Netherlands: Netherlands National Committee for the IAH. Pp. 3-18. Available at http:// siteresources.worldbank.org/INTWRD/Resources/GWMATE_Final_booklet.pdf (accessed on June 7, 2013). Wandert, B. 2009. Fergana Valley Water Resources Management project- Phase I. Environmental assessment. The World Bank.141p. Wang, J.; Huang, J.; Rozelle, S. 2005. Evolution of tubewell ownership and production in the northern China plain. Austral. J. Agric. Res. Econ. 49(2): 177-195. Waterloo Hydrogeologic Inc. 2000. Visual MODFLOW user’s manual. Waterloo Hydrogeologic Inc., Canada. Xu, G.; Liu, L.; Fei, Y.; Li, D.; Qian, Y. 2009. Research on the adjustment of groundwater storage in the North China Plain. Resources Science 31: 375-381 (in Chinese). 43 IWMI Research Reports 151 Managed Aquifer Recharge: The Solution for Water Shortages in the Fergana Valley. Akmal Karimov, Vladimir Smakhtin, Aslon Mavlonov, Vecheslav Borisov, Inna Gracheva, Fazleddin Miryusupov, Jamol Djumanov, Tatyana Khamzina, Rustam Ibragimov and Botir Abdurahmanov. 2013. (Also available in Russian). 150 Glacier Systems and Seasonal Snow Cover in Six Major Asian River Basins: Hydrological Role under Changing Climate. Oxana S. Savoskul and Vladimir Smakhtin. 2013. 149 Glacier Systems and Seasonal Snow Cover in Six Major Asian River Basins: Water Storage Properties under Changing Climate. Oxana S. Savoskul and Vladimir Smakhtin. 2013. 148 Evaluating the Flow Regulating Functions of Natural Ecosystems in the Zambezi River Basin. Matthew McCartney, Xueliang Cai and Vladimir Smakhtin. 2013. 147 Urban Wastewater and Agricultural Reuse Challenges in India. Priyanie Amerasinghe, Rajendra Mohan Bhardwaj, Christopher Scott, Kiran Jella and Fiona Marshall. 2013. 146 The Water Resource Implications of Changing Climate in the Volta River Basin. Matthew McCartney, Gerald Forkuor, Aditya Sood, Barnabas Amisigo, Fred Hattermann and Lal Muthuwatta. 2012. 145 Water Productivity in Context: The Experiences of Taiwan and the Philippines over the Past Half-century. Randolph Barker and Gilbert Levine. 2012. 144 Revisiting Dominant Notions: A Review of Costs, Performance and Institutions of Small Reservoirs in Sub-Saharan Africa. Jean-Philippe Venot, Charlotte de Fraiture and Ernest Nti Acheampong. 2012. 143 Smallholder Shallow Groundwater Irrigation Development in the Upper East Region of Ghana. Regassa E Namara, J.A. Awuni, Boubacar Barry, Mark Giordano, Lesley Hope, Eric S. Owusu and Gerald Forkuor. 2011. 142 The Impact of Water Infrastructure and Climate Change on the Hydrology of the Upper Ganges River Basin. Luna Bharati, Guillaume Lacombe, Pabitra Gurung, Priyantha Jayakody, Chu Thai Hoanh and Vladimir Smakhtin. 2011. 141 Low-cost Options for Reducing Consumer Health Risks from Farm to Fork Where Crops are Irrigated with Polluted Water in West Africa. Philip Amoah, Bernard Keraita, Maxwell Akple, Pay Drechsel, R.C. Abaidoo and F. Konradsen. 2011. Electronic copies of IWMI's publications are available for free. Visit www.iwmi.org/publications/index.aspx Postal Address P O Box 2075 Colombo Sri Lanka Location 127 Sunil Mawatha Pelawatta Battaramulla Sri Lanka Telephone +94-11-2880000 Fax +94-11-2786854 E-mail iwmi@cgiar.org Website www.iwmi.org ISSN: 1026-0862 ISBN: 978-92-9090-770-1