INTRODUCTION
Problem Statement
The VGTB basin faces various issues that can be viewed from both specific and general perspectives Similar to other rivers in Vietnam, integrated watershed management in VGTB is hindered by overlapping state administration, resulting in inefficiencies in water resources planning strategy The involvement of multiple ministries, including the Ministry of Natural Resources and Environment, Ministry of Agriculture and Rural Development, and Ministry of Construction, in managing river-related assets leads to confusion and low applicability of water allocation planning studies This fragmented approach exacerbates issues such as overexploitation, water quality degradation, and flow regime changes, making it challenging to address these problems comprehensively.
The involvement of stakeholders, particularly local communities, in water resource allocation planning is often overlooked, leading to inefficiencies in management In many basins, committees established to oversee resource management have proven ineffective, and the lack of coordination between administrative counties has hindered proficiency The rapid development of industrial parks, dams, and urbanization has resulted in increased hazardous waste, pollution, and degradation of coastal areas, ultimately giving rise to conflicts over downstream water allocation A notable example of these challenges is the VGTB River basin, where the regulation of reservoirs poses a significant problem, with four large hydropower projects and over 820 irrigation works, including 72 reservoirs, 546 spillways, and 202 pumping stations, further complicated by planned hydropower projects on the mainstream of Vu Gia.
The Thu Bon River Basin had proposed 10 hydropower plants with a total capacity of 1,200 MW by 2020, but studies have shown severe impacts of reservoirs on the area, particularly in terms of inundation and drought The natural flooding patterns have become more extreme and unpredictable due to human activities upstream, while irrational water management in the reservoirs has caused salinity intrusion and inundation in downstream areas, such as the 2012 incident at Han estuary and the 2009 flooding in Quang Nam Furthermore, the use of reservoirs has deviated from their original design, prioritizing electricity generation over flood control, leading to man-made and flash floods in the downstream Despite the Prime Minister's issuance of a reservoir operation process in 2010, the safety of downstream citizens remains at risk, as evident from the series of hydro-flood incidents that have resulted in loss of property and lives.
Recent extreme weather events, such as the 2009 and 2013 storms, have highlighted the importance of accurate water availability assessments However, the reliability of data on water storage in reservoirs is limited, making it challenging to predict water levels using traditional models This study assumes that the flood discharge process adheres to established protocols, aiming to improve the accuracy of water availability assessments and mitigate the impacts of extreme weather events.
The VGTB river basin plays a vital role in the socioeconomic development of the Central Coast, particularly in Quang Nam and Da Nang provinces, by providing a crucial source of water for various development needs The VGTB River system supports over 45,000 hectares of agricultural and domestic production, catering to nearly 2 million people in the basin, while also possessing significant hydropower potential Notably, the Vu Gia River, as it flows through Da Nang city, is instrumental in the city's socioeconomic development, supplying approximately 75 million cubic meters of raw water to urban water treatment plants and over 100 million cubic meters for agricultural purposes annually.
The river plays a crucial role in supporting economic activities and livelihoods, while also regulating the climate and creating breathtaking landscapes, particularly around the Han estuary However, rapid industrialization and development in this key central economic region have put immense pressure on the basin's water resources, especially during dry seasons The construction of dams upstream has exacerbated water scarcity downstream, leading to conflicts among water users and highlighting the need for reallocation Currently, the majority of water resources are allocated to agriculture, despite industry offering higher water yields, resulting in decreased total benefits from industrial production Therefore, optimizing the VGTB stream water allocation is vital to ensure sustainable development and maximize the benefits of water resources in the region.
To ensure integrated and effective management, a study on resource allocation is crucial, taking into account the basin's characteristics and management aspects This approach enables the consideration of both technical factors and business efficiency, ultimately leading to optimal resource utilization By leveraging tools like SWAT, decision-makers can establish a linkage between resource allocation and management, facilitating informed decisions that balance technical and economic aspects.
The Linear Programming (LP) approach can be effectively utilized to compute allocation based on the Integrated Water Resources Management (IWRM) framework, taking into account the key components of the hydrological cycle By considering the advocacy process of water within the basin, LP can facilitate informed decision-making Additionally, this approach enables the incorporation of crystal economic efficiency, ensuring optimal allocation of water resources.
Objectives of Study
The overall objective of this study is to propose an optimal water allocation plan in the Vu Gia - Thu Bon River basin The specific objectives are as follows:
To calculate the total allocable water availability in the VGTB river basin;
To identify the water demands of sectors and water prices in the basin;
To build and mathematically solve the objective function and constraints towards target of the study.
Scope of Study
The study focuses on the following issues:
Overview of previous studies on water allocation planning and linear programming;
Application of hydrological model to calculate the water availability in the study basin;
Application of linear programming to specify a water allocation mechanism maximizing the revenue of supplier from the total available water volume.
Research Questions
The primary objective is to develop an optimal allocation mechanism for a limited water supply, with the ultimate goal of maximizing benefits for the supplier To achieve this, the study aims to address key questions, providing a comprehensive framework for allocating this scarce resource effectively.
How much water is available to allocate in the study area?
Which method is used to assess the allocable water availability in the study area? And how to utilize this method?
How much water is required by sectors up to next five years basing on national standard?
What is the highest number of earnings that water supplier can obtain from accessible water allocated to sectors?
Vu Gia – Thu Bon River Basin
Vu Gia - Thu Bon River system is located in the Central Coast Region of Vietnam with
10350 km2 total basin area, of which majority is belonged to Quang Nam Province and
Da Nang City while a small part is administrated by Kon Tum Province with 301.7 km 2
VGTB River basin (16 o 03’- 14 o 55’ N; 107 o 15’- 108 o 24’ E) is bounded on the North by
The Cu De river basin lies to the north, while the Tra Bong and Se San river basins border the region to the south To the west, the area is bounded by Laos, and to the east, it is flanked by the East Sea and the Tam Ky river basin.
Figure 1.1: Vu Gia – Thu Bon river basin
The Vu Gia - Thu Bon (VGTB) river basin spans across 17 administrative districts and cities, encompassing Bac Tra My, Nam Tra My, Tien Phuoc, Phuoc Sơn, Hiep Duc, Dong Giang, Tay Giang, Nam Giang, Que Son, Duy Xuyen, Dai Loc, Dien Ban, Hoi An, Da Nang, Hoa Vang, and parts of Thang Binh and Dak Glei, covering the provinces of Kon Tum, Quang Nam, and Da Nang City.
The topography of the VGTB river basin is strongly fragmented and inclined west to east, forming four main categories of terrain as follows:
The basin's landscape is dominated by mountainous terrain, with the Truong Son Mountains covering a significant portion of the area, featuring elevations ranging from 500m to 2000m Notably, the region is bounded by prominent peaks, including Mang Mountain at 1768m and Ba Na Mountain at 1467m, which reach heights of 1000m to 2000m, defining the basin's rugged topography.
Notable mountains in the region include A Tuat, Lum Heo, and Tien, situated upstream of the Vu Gia River, as well as Ngoc Linh and Hon Ba, located upstream of the Tranh River These mountain ranges originate from Hai Van Pass in the north and curve westward, southwestward, and then southward, forming a bow-shaped structure around the basin This unique geography makes the basin prone to capturing Northeast monsoon winds and weather patterns from the East Sea, resulting in heavy rainfall, flash floods in mountainous areas, and inundation in low-lying regions.
The region's hilly terrain, situated east of the mountainous area, features rounded or fairly flat peaks with a slope of approximately 20 to 30 degrees Characterized by a gradual decrease in elevation from west to east, this area stretches from Tra My District in the north to the western border of Duy Xuyen District Notably, it serves as a confluence for several significant tributaries of the Thu Bon River, including the Tranh, Truong, Tien, Lan, Ngon, Khe Dien, and Khe Le rivers.
Lowland terrain: Elevation of plains in the VGTB river system is lower than 30 m with relatively flat and homogeneous terrain, concentrating mainly on the East of the basin.
The narrow plain in the region, stretching in a North-South direction, is a result of its proximity to the coastal mountains Formed by the convergence of ancient alluvial sediment and silt deposits from the sea, rivers, and streams, this lowland terrain covers several districts, including Dai Loc, Duy Xuyen, Dien Ban, Thang Binh, Hoi An, Tam Ky, and Hoa Vang Notably, the area is also home to several small rivers, such as Khe Cong, Khe Cau, and Quang Hue.
Coastal sand terrain is characterized by vast expanses of sand dunes that originate offshore and are driven ashore by wind, predominantly from the west This natural process results in the formation of hundreds of kilometers of wavy sand dunes that stretch along the coast, creating a unique and dynamic landscape.
1.5.3 Rainfall Characteristics in the Dry Season
The dry season in the Vu Gia-Thu Bon River basin typically lasts from January to August, accounting for approximately 30% of the total annual rainfall The three months with the lowest rainfall density are February, March, and April, with February experiencing the most significant reduction in the Vu Gia River basin and March in the Thu Bon River basin, each contributing a mere 1% to the total annual rainfall.
Table 1.1: Rainfall in the dry season, the three-lowest-month and the lowest month (mm )
Season Three-lowest-month Lowest month
Season Three-lowest-month Lowest month
The dry season in the basin coincides with peak agricultural production periods, encompassing the winter-spring harvest from January to April and the mid-year-fall crop from May to September, significantly impacting the region's water supply capacity, particularly during the heightened water demand from January to May.
The dry season in the territory typically lasts from January to September each year, with the lowest runoff usually occurring in April However, if additional rainfall does not occur in May and June, the minimum runoff is typically recorded around that time, indicating a prolonged period of reduced water flow.
For rivers spanning basin territories over 300km², the lowest stream flow typically occurs in April, whereas smaller basins under 300km² experience the lowest runoff during the summer months, specifically between June and August.
Table 1.2: Low-flow characteristics of the VGTB River
Thanh My - Vu Gia Nong Son - Thu Bon (F=1,850 km 2 ) (F=3,155 km 2 )
Time of occurrence Jan - Aug Jan - Aug
Time of occurrence Feb - Apr Mar - May
Time of occurrence April 1983 June 1998
The low flow is depended on groundwater reserves and rainfall density in the basin The dry season can be divided into two periods:
- Stable flow: During this period, flow is mainly fed by volume of water reserved in the river, causing a chronologically decreasing trend and then stability (fromJan to Apr annually).
- Instable flow: From May to July, water supplied to the flow is not only from groundwater but additional rainfalls.
Due to this characteristic, the lowest flow usually happens twice in the rivers around March to April and June to July.
Figure 1.2: Mean flow in the dry season of 1981-2010 periods
Figure 1.3: Low-flow module (Source: Water Resources Investigation and Assessment of VGTB River Basin Project)
I II II IV V VI VII VIII IX
The low runoff in the river accounts for 40-45% of the total annual flow, with the most significant decrease typically occurring in the upstream territories during the dry season, where the mean stream module fluctuates between 30-40l/s.km² Notably, the Northern and Northwestern parts of Quang Nam areas, particularly the Bung and Kon River basins, record the lowest runoff, with the low-stream module dropping as low as 10l/s.km² in these regions.
Table 1.3: The lowest flow characteristics in the basin
Station F lv (km 2 ) Q k,tb (m 3 /s) Cv Cs
Table 1.4: The lowest flow at some main locations in the river basin
Station River F (km 2 ) M min-month
Nong Son Thu Bon 3.150 8,6 VI/1998 4,63 17/8/1977
LITERATURE REVIEW
Water Allocation Planning
Effective water allocation planning is crucial in addressing the global challenge of water scarcity, which threatens sustainable development, food security, and ecosystem health As a sharing methodology of limited water resources, water allocation must provide discerning solutions to questions of deliberation and insurance between topographical regions and water users A comprehensive approach to water allocation planning is essential in avoiding conflicts related to water use interest at multiple scales and maintaining a healthy ecosystem.
The general objective and particular goals of water allocation planning have undergone significant changes over time, closely tied to the human development index In contrast to previous methodologies, modern water allocation mechanisms have become increasingly complex, taking into account multiple perspectives This approach, as outlined by Robert Speed et al (2013), involves two key steps: assessing available water for allocation and determining an allocation mechanism that meets the demands of various sectors The late 20th century saw a series of pivotal events that led to the publication of influential documents, shaping modern water management, including the seminal Brundtland Report.
The concept of sustainable development was introduced in 1987, followed by the Dublin Principles in 1992, which established four key principles as the foundation of Integrated Water Resources Management (IWRM) Agenda 21, an action plan stemming from the 1992 United Nations Conference on Environment and Sustainable Development in Rio de Janeiro, further defined IWRM, solidifying its importance in managing water resources effectively.
Recognizing water as a vital component of the ecosystem, a natural resource, and a social and economic good, the United Nations Department of Economic and Social Affairs (UNDESA) emphasizes its crucial role in determining the nature of its utilization based on its quantity and quality Effective water management is essential to address ecosystem degradation and boost the efficiency of economic activities, which are often hindered by inadequate water management practices.
Figure 2.1: Basin water allocation agreements and plans in the twentieth century (Robert Speed et al, 2013)
The shift towards complex water allocation planning frameworks is often a response to intensifying competition and scarcity of basin water resources Notably, the severe environmental crisis in the Murray-Darling basin in the early 1990s led to significant changes in the Murray-Darling Agreement and the implementation of basin-scale regulation In regions like Western Australia, water abstraction is managed through individual licenses based on collective or geographic water allocation guides, which set out abstraction limits and management strategies for the present and future Similar approaches to water allocation planning can be seen in the Colorado River basin, where water sharing is structured by key agreements, including the landmark 1922 Colorado River Compact.
The compact in question exemplifies a basic water allocation mechanism between regions, but its inflexibility has raised concerns Notably, it fails to accommodate annual adjustments, neglects to incorporate environmental flow considerations, and lacks a temporal regulation mechanism to respond to fluctuations in climate, water demand, and priority shifts.
Figure 2.2: Water allocation planning model in Western Australia
In Asia, the transboundary nature of many river basins has led to the establishment of international river basin management institutions, promoting cooperation and conflict prevention among nations A notable example is the Indus River treaty between India and Pakistan, which allows India to utilize the water from three upstream tributaries while allocating the remaining volume to Pakistan The 1991 Water Accord, signed by Pakistani state chief ministers, has successfully implemented a water allocation mechanism, incorporating measures that respond to seasonal variations and environmental flow However, the allocation process has limitations, focusing only on base scenarios and neglecting alternative water supply sources, while the environmental minimum flow allocation lacks clear definition, potentially threatening ecosystem vulnerability.
In Vietnam, the evolution of river water allocation planning can be categorized into three distinct periods: pre-2008, 2008-2013, and post-2013 Prior to 2008, the country relied on the 1998 Law on Water Resources to guide irrigation planning, which was typically divided into three categories: comprehensive planning, single-sector planning, and bilateral planning Comprehensive planning involved developing and arranging activities at a national or large-scale level, impacting socioeconomic and natural development Single-sector planning focused on individual water use sectors, such as urban water supply or irrigation system planning, at a sub-regional or local scale Bilateral planning, on the other hand, involved coordinating water use plans across sectors, including land use, irrigation, transport, and rural planning, and was often more complex and comprehensive than single-sector planning.
During this period, integrated plans are only passed by competent authorities without formal written approvals.
In 2008, the Vietnamese Government introduced Decree No 120/2008/ND-CP, outlining river basin management regulations, including three key aspects of water resources planning: allocation, protection, and prevention of water-related harm The Ministry of Natural Resources and Environment further solidified these regulations with Circular No 15/2009/TT-BTNMT in 2009, detailing the economic-technical norms and procedures for water resources planning, encompassing five main components: surface and groundwater allocation and protection planning, as well as prevention and remedy of water-related harm Notably, provincial water resources planning requires approval from the Chairman of the respective province or centrally run city, following stakeholder consultation, as stipulated by the Law on Water Resources and Decree No 120/2008/ND-CP.
The Law on Water Resources No 17/2012/QH13, effective from January 1, 2013, has introduced key regulations on water resources planning According to Article 15, water resources planning encompasses national, inter-provincial river basin, and provincial levels, including plans for provinces and centrally run cities The authority responsible for approving these plans is outlined in Article 21, with provincial People's Committees developing plans for submission to the People's Councils, following written opinions from the Ministry of Natural Resources and Environment The planning process adheres to technical-economic norms issued by the Ministry of Natural Resources and Environment and Circular 05/2013/TT-BKH by the Ministry of Planning and Investment, marking a departure from previous regulations.
Figure 2.3: Water resources planning framework in Vietnam
The "Water resources planning in Dong Nai Province to 2020" serves as a prime example of applying the framework outlined in Circular No 15/2009/TT-BTNMT This provincial-scale plan focuses on optimizing water resource exploitation and utilization, safeguarding river and water source integrity, and proactively mitigating degradation and depletion By doing so, the plan aims to address adverse water-related consequences and support socioeconomic development in Dong Nai Province, with implementation divided into two distinct phases.
2012 to 2015 and the second four-year period from 2014 to 2020 with concrete doings:
Effective water resource management in Dong Nai Province involves a three-pronged approach, encompassing planning for the allocation of surface water and groundwater, protection of these resources, and prevention and mitigation of the consequences of water-related harm This comprehensive strategy is closely aligned with the region's socioeconomic development plan, land use plan, and urban water supply and industrial zone plans up to 2020 By integrating these plans, Dong Nai Province aims to ensure sustainable water resource management that supports its overall development goals.
A notable example of water allocation planning in Vietnam is the "Water Resources Allocation Planning in Lang Son Province to 2020, Orientation to 2030" study This initiative aimed to assess the current state of water resource management, exploitation, and utilization in the province, while proposing sustainable solutions for water resource exploitation and use The study adhered to four key allocation principles, analyzing three scenarios to achieve its objectives The principles prioritized sectors generating the highest economic benefits, while ensuring adequate water allocation for domestic use and promoting water security levels, with a focus on sharing benefits among sectors and addressing insufficient water supply for production.
In the Dong Nai case study, water resources planning solutions prioritize design water supply security levels, where sectors with lower security levels must accept the associated risks The current allocation rate is followed, allocating remaining water to sectors after domestic use needs are met All sectors must adjust their water needs to align with the allocation mechanism in the event of water shortages Additionally, prioritization is given to objectives that serve political and social stability, as well as poverty alleviation, particularly in regions or sectors receiving preferential policies to maintain social security.
Soil and Water Assessment Tool (SWAT)
2.2.1 Historical Development of SWAT Model
The Soil and Water Assessment Tool (SWAT) is an ongoing development project at the USDA Agricultural Research Service (ARS), spanning nearly 40 years Originating from the "Simulator for Water Resources in Rural Basins" (SWRRB) model, SWAT was designed to simulate water systems and sediment transport in ungauged basins across the United States The SWRRB model evolved from the early 1980s' CREAMS hydrologic model modification, which later influenced the development of the Routing Outputs to Outlet (ROTO) model in the 1990s.
Figure 2.5: Water Resources Allocation Planning in Lang Son Province
The ROTO model was developed to aid in the administration of the underground stream in the Indian Fields of Arizona and New Mexico, covering a vast area of several thousand square kilometers Commissioned by the US Bureau of Indian Affairs, this model plays a crucial role in managing the region's water resources.
A significant milestone in the development of simulation models was the integration of the SWRRB and ROTO models into a single, unified model - the SWAT model Notably, the SWAT model retained all the options of the SWRRB model, making it a highly versatile and valuable tool for simulating processes across vast areas.
The SWAT model has undergone significant transformations since its initial introduction, with notable revisions in versions 94.2, 96.2, 98.1, 99.2, and 2000, as documented by Arnold and Fohrer (2005) and Neitsch et al (2005) Today, the SWAT model is a sophisticated, physically-based model that operates on a daily time step, effectively simulating water flow, sediment circulation, and chemical transport in ungauged watersheds Its computational efficiency enables the performance of extended simulations, making it a valuable tool in hydrological modeling The model discretizes the catchment into multiple sub-catchments, which are further divided into homogeneous hydrologic response units (HRUs) based on land use, vegetation, and soil characteristics.
The SWAT model relies on a range of essential inputs to function effectively, including daily rainfall, maximum and minimum air temperature, solar radiation, relative air humidity, and wind speed These inputs can be sourced directly from metering stations or pre-computed beforehand, providing the necessary data to drive the model's simulations.
The Green-Ampt infiltration method is a widely used approach for applying daily measured or generated rainfall data, originally developed by Green and Ampt in 1911 This method takes into account snowfall, which is determined based on precipitation levels and mean daily air temperatures By utilizing maximum and minimum daily air temperatures in its computations, the Green-Ampt model provides a comprehensive framework for analyzing infiltration patterns.
Application of climate inputs includes the following: (1) up to ten elevation zones are simulated for calculation of rainfall distribution per elevation and/or snowmelt process,
(2) climate inputs are adapted to simulation model requirements, and (3) forecast of weather conditions is performed as a new option of the SWAT 2005.
A full hydrologic equalization for each Hydrologic Response Unit (HRU) involves a comprehensive process that aggregates and accounts for evaporation from plants, determines significant rainfall and snowmelt, and simulates the interaction between surface flow and the soil layer This process also encompasses water infiltration into deeper layers, evapotranspiration, and the movement of subsurface streams and underground flows Ultimately, the goal is to accurately model water accumulation and distribution within the HRU, providing a detailed understanding of the hydrologic cycle and its various components.
The SWAT model offers two estimation options for surface runoff from Hydrologic Response Units (HRUs), utilizing either the daily or hourly precipitation combined with the USDA Natural Resources Conservation Service (NRCS) curve number (CN) method or the Green-Ampt method The CN technique effectively accounts for water retention on plants, while the Green-Ampt method accurately replicates explicit water retention Additionally, the model calculates water accumulation in soil and its flow lag through the processes of water redistribution between soil layers.
Sub-surface stream simulation has been effectively demonstrated in Arnold et al (2005) for specific soil classes with fissures Furthermore, SWAT 2005 offers enhanced capabilities, including new options for simulating water level fluctuations in soil on Hydrologic Response Units (HRUs) with intermittent flows, providing more accurate modeling of complex hydrological processes.
Estimating potential and actual evapotranspiration relies on three primary routines: the Penman-Monteith, Priestly-Taylor, and Hargreaves methods Water exchange between the soil and deeper layers occurs through the sub-surface soil layer, driven by water not utilized by plants or evaporated This sub-surface stream contributes to subsurface supplies, ultimately feeding into deeper repositories Water that infiltrates these deepest repositories is considered lost to the system, effectively becoming a system yield.
2.2.2 Theoretical Base and Applications of SWAT Model
The SWAT model allows for the simulation of various physical courses of action within a basin by partitioning a catchment into sub-catchments, enabling the analysis of distinct attributes of vegetation, soil, and hydrologic processes This division facilitates the identification of significant catchment regions and their characteristics Input data for each sub-catchment is categorized into climate, hydrologic response units (HRUs), reservoirs/lakes, groundwater, stream network, and catchment runoff HRUs, typically square-shaped areas within sub-catchments, are characterized by homogenous vegetation, soil, and land use classes, allowing for a more detailed understanding of hydrologic responses.
The foundation of the technique lies in the water balance of the catchment area, regardless of the issue being demonstrated and investigated by the model To accurately predict the movement of pesticides, sediment, or nutrients, the model simulates the hydrologic cycle, integrating overall water flow in the catchment area Hydrologic simulations in the catchment territory can be broadly categorized into two stages: the soil phase, where surface and subsurface soil interactions occur, and the movement of sediment, nutrients, and pesticides through water streams in all sub-catchments.
Hydrologic cycle is simulated by SWAT model, which is based on the following balance equation:
The soil's moisture levels can be calculated using the equation SWt = SW0 + Rday - Qsurf - Eai - Wseep + Qgw, where SW0 represents the base humidity of the soil, SWt is the humidity at a given time, Rday is the rainfall volume, Qsurf is the surface runoff, Eai is the evapotranspiration, Wseep is the seepage of water into deeper layers, and Qgw is the underground runoff.
Figure 2.6: Balance scheme of SWAT model
The SWAT model relies on a comprehensive set of climate and hydrologic inputs, including rainfall, air temperature, solar radiation, wind speed, relative air humidity, snow pack, snowmelt, and elevation zones, as well as water volume on plants, infiltration, and water seepage into deeper soil layers Additionally, the model incorporates evapotranspiration, sub-surface flow, surface flow, and the presence of lakes and river networks, along with underground flow Furthermore, it also takes into account factors related to vegetation growth and development, erosion on the catchment area, nutrients, pesticides, and land use, providing a holistic approach to hydrologic modeling.
The SWAT model is a physically-based approach that utilizes a water balance framework, comprising five linear reservoirs as illustrated in Figure 1 Each reservoir is governed by a set of applied equations that quantify water balance and connections between stores, representing possible water pathways, including both surface and subsurface flows.
Figure 2.7: Scheme of linear repositories in SWAT model
Linear Programming
Researchers have explored various methods to determine the optimal operating mechanism for reservoir chains, particularly those constructed on the Firat River in Turkey for water supply and power generation purposes A notable study by Dagli and Miles (1980) applied simulation, linear programming, and optimal random methods to address operational challenges Similarly, Dandy and Crawley (1992) investigated the application of linear programming in reservoir system planning and operation, further contributing to the development of efficient management strategies.
A pioneering model developed by Tejada et al in 1995 optimized the operation of hydropower plants by addressing the uncertainty of random hydrological inputs and fluctuating electricity demands This innovative approach employed dynamic programming to analyze hydrological sequences using various methods, including monthly averages, frequency distributions, and Markov chains By incorporating variable power demand and reasonable penalty costs for insufficient power supply, the model provided a comprehensive framework for efficient hydropower plant operation, as demonstrated by its successful application to the Shasta-Trinity system in California, USA.
Optimization models for water resources management in river basins have been extensively studied and developed to demonstrate the effective application of optimal algorithms in river basin water management A notable example is the work of Lee and Howitt (1996), who developed models for the Colorado River Basin to determine the optimal benefits of water supply for irrigation, domestic, and industrial production while minimizing saltwater intrusion Their analysis of three alternatives revealed that prioritizing economic benefits alone would lead to water transfers from agriculture to domestic and productive sectors due to high economic efficiency, while incorporating supportive measures to control salinity intrusion, with either fixed or flexible plant structures, resulted in significant reductions in saltwater intrusion.
Researchers Ximing Cai et al (2001) introduced a comprehensive model that integrates the economy, agriculture, and hydrology for effective river basin management This model provides a general framework for managing river basins, emphasizing agriculture as the primary water-consuming sector Notably, the model considers saline intrusion caused by irrigation as a significant environmental impact factor By combining all components into a single, closed model, the framework utilizes a simple yet effective decomposition method to solve the integrated management problem.
Approach The model was applied to the actual case in the Darya River basin in Central Asia.
Researchers Ito et al (2001) introduced a Decision Support System (DSS) for river basin water management, integrating a hydrological cycle simulation model and a risk assessment model to provide a comprehensive approach This synthetic model was successfully applied to a real-world scenario in the Chikugo river basin system, which features multi-purpose reservoirs.
The Optimal Economic-Technique Model, developed by Richard E Howitt et al in 1999, demonstrates the feasibility and practicality of applying water resources optimization models on a large scale to achieve economic goals This pioneering model has been successfully applied in real-world water management scenarios, notably in California, showcasing its effectiveness in optimizing water resources By integrating economic objectives into water management decision-making, the model provides a valuable framework for efficient and sustainable water allocation.
APPLICATION OF SWAT
Input Data Processing
Input data comprising DEM and Land cover are downloaded from official website of U.
Figure 3.2: Screen shot of official website of USGS
The USGS offers free access to the SRTM 1 Arc-Second Global elevation data, providing worldwide coverage with a resolution of 1 arc-second (30 meters) Released in phases since September 24, 2014, this high-resolution global data set is available for open distribution Primarily intended for scientific applications, the SRTM elevation data is best utilized with Geographic Information System (GIS) software or other specialized applications.
The USGS offers a valuable resource in the form of 0.5 km MODIS-based Global Land Cover Climatology data, which provides detailed information on land cover types This dataset is derived from 10 years of Collection 5.1 MCD12Q1 land cover type data, spanning from 2001 to 2010 The resulting map is generated by selecting the land cover classification with the highest overall confidence for each pixel, offering a comprehensive and reliable representation of global land cover patterns.
2010 (Broxton et al., 2014) As such, they are reflective of the training data for the
The MDC12Q1 data has undergone re-gridding from the MODIS sinusoidal grid to a regular latitude-longitude grid, resulting in a high-resolution map with 43,200x86,400 pixels, equivalent to 15 arc seconds Notably, this dataset exhibits limitations near the edges of the map, specifically within 0.05 degrees of 180 degrees longitude and over parts of Antarctica, primarily south of -85 degrees latitude.
Figure 3.3: Screen shot of MODIS-based Global Land Cover Climatology
Soil data were sourced from the Food and Agriculture Organization's (FAO) official website, utilizing the FAO-UNESCO Soil Map of the World as the primary dataset The Digitized Soil Map of the World, scaled at 1:5,000,000, was projected geographically in latitude and longitude, and intersected with a template that included water-related features such as coastlines, lakes, glaciers, and double-lined rivers.
Data are processed in ArcSWAT which is an ArcGIS-ArcView expansion and graphical user input interface for SWAT Simulation of SWAT model can be depicted as figure
17 The procedure is started by delineating sub-watersheds in light of a programmed system utilizing DEM information Consequently, land use, soil and slope characterization for a watershed is performed utilizing summons from the HRUAnalysis SWAT model obliges area utilize and soil information to focus the zone and the hydrologic parameters of every area soil classification simulated inside every sub- catchments ArcSWAT likewise permits the integration of area slope classes when characterizing hydrologic response units.
Figure 3.4: Screen shot of FAO official website
Figure 3.5: SWAT Model Simulation (Source: NASA-CASA Project)
The recently released report provides a comprehensive overview of land use, soil, and slope class distribution within the watershed and its sub-watershed units, following the completion of the overlay process.
Incorporating stream flow subsidence data into base flow estimations can significantly enhance their accuracy This is because the natural stream flow gauge in the stream can be accounted for by stream flow recession, ultimately leading to more precise base flow calculations.
Table 3.1: Information of basin after overlay
A stream's base flow is significantly influenced by its retreat period, with shorter periods resulting in higher variability of occasional base flow impact Understanding the occasional components of direct stream and base flow is crucial for effective waterway management, and stream flow retreat data can provide valuable insights The alpha factor, or base flow-retreat coefficient, is a key parameter in the SWAT model, but its accuracy can be compromised by other parameters related to stream flow Consequently, precise base flow evaluations can be challenging to achieve, and uncertainties in stream flow forecasts can propagate into the accuracy of base flow estimates at ungauged watersheds Therefore, stream flow predictions made using SWAT must accurately reflect the retreat period by considering the alpha factor to ensure reliable base flow estimates.
The alluvial fields in the region are predominantly used for harvesting crops such as rice, sugarcane, and tobacco, while sloping territories are more suited for tea, rubber, and pepper plantations Historically, the area was once heavily forested, but widespread deforestation has occurred to make way for food and industrial crops By 2005, the basin's forest area had increased to 445,748 hectares, accounting for 43.5% of the region, with 405,050 hectares of natural forest and 40,698 hectares of plantations However, despite this growth, the basin's water storage and control capabilities remain inadequate, leading to soil erosion, surface water and groundwater depletion, and increased sedimentation downstream.
Sub-catchments Delineation
Figure 3.8: Sub-catchments divided by SWAT model
Figure 3.9: Final sub-catchments map
The final sub-catchment map is delineated based on natural features, topographic divisions of rivers and tributaries, and corresponding administrative boundaries and management units, taking into account the characteristics of the water system to facilitate effective management and exploitation of water resources, as well as the needs and characteristics of water use and supply, including drainage direction after use, all of which are accurately mapped using ArcGIS technology.
The entire 10.350 km2 basin area is devided into 5 sub-basins as follows:
Table 3.2: Sub-basins of VGTB basin
Sub-basin Area (km 2 ) Administrative territory
Vu Gia upstream 5,242.46 Districts of Tay Giang, Dong Giang, Nam Giang and part of
Dai Loc, Phuoc Son, Dak Glei (Kon Tum)
Thu Bon upstream 3,215.43 Districts of Ba Tra My, Nam Tra My, Tien Phuoc, Hiep Duc,
Nong Son and part of Phuoc Son, Dai Loc, Duy Xuyen
Ly Ly river 373.92 Districts Que Sơn and part of Thang Binh
Tuy Loan river 421.73 Districts of Thanh Khe, Hai Chau, Son Tra and Hoa Vang
VGTB downstream 1,096.46 Hoi An City, Districts of Dien Ban, Ngu Hanh Son, Cam Le and part of Dại Loc, Duy Xuyen.
Reservoir Processing
To accurately model the study area's electricity generation needs, reservoirs are designated to store a specific monthly volume of water This input is crucial, as reservoirs significantly impact natural runoff patterns The water volume changes are reflected in the output nodes of the sub-basins, which were previously delineated Notably, three key hydropower reservoirs are included in this study: A Vuong, Dak Mi, and Tranh.
Some important parameters declared for the reservoir is defined as follows:
Table 3.3: Definitions of reservoir parameters
MORES Month of the reservoir became operational (0-12).
If 0 is input for MORES, SWAT model assumes the reservoir is in operation at the beginning of the simulation
IYRES Year the reservoir became operational.
If 0 is input for IYRES, SWAT model assumes the reservoir is in operation at the beginning of the simulation
RES_EVOL Volume of water needed to fill the reservoir to the emergency spillway (10 4 m 3 )
RES_PVOL Volume of water needed to fill the reservoir to the principle spillway (10 4 m 3 ) RES_VOL Initial reservoir volume
The initial reservoir volume is determined by the reservoir's status at the start of the simulation period, with the volume on the first day of simulation serving as the initial volume if the reservoir already exists Conversely, if the reservoir commences operation during the simulation, the RES_VOL value represents the reservoir's volume on the day it becomes operational, which is 10,000 cubic meters (10^4 m^3).
The RESMONO value represents the monthly reservoir outflow, calculated from a file containing the average daily flow rate for each month of the reservoir's operation In conjunction with this, the STARG (mon) metric provides the monthly target reservoir storage, expressed in units of 10^4 cubic meters.
According to Decision No 909/QD-TTg dated June 16, 2014, issued by the Prime Minister, specific parameters govern the operation of reservoirs in the Vu Gia - Thu Bon basin during the annual flood season From September 1 to December 15 each year, the A Vuong, Dak Mi 4, and Song Tranh 2 reservoirs are operated based on a priority principle.
To guarantee the absolute safety of hydropower works such as A Vuong, Dak Mi 4, and Song Tranh 2, water elevation is strictly controlled to not exceed the maximum allowable water elevation, particularly during floods with a return period of 1000 years or less.
- To make a contribution reducing flood in the downstream;
- To ensure electricity generation efficiency.
Figure 3.11: Edit Reservoir Parameters Table
Table 3.4: Technical parameters of reservoirs
Parameters A Vuong Dak Mi 4 Song Tranh 2
Total reservoir volume 343.55 mil m 3 312 38 mil m 3 729.20 mil m 3 Effective reservoir volume 266.48 mil m 3 158.26 mil m 3 521.10 mil m 3 Dead reservoir volume 77.07 mil m 3 154.12 mil m 3 208.10 mil m 3
First electricity generation day – unit 1 26/9/2008 31/12/2010 6/2011First electricity generation day – unit 2 19/12/2008 28/2/2011 12/2011
Land Cover Scenario
The study utilized the SWAT model for the base scenario in 2001 and incorporated the Land Use Update Edit tool to modify land cover percentages, specifically focusing on the conversion of 10% of agricultural land to urban and industrial land by 2020, thereby assessing the impacts of land use changes on the environment.
Figure 3.12: Land Use Update Edit tool
The calibration and validation process of model is performed by using SUFI-2 withinSWAT-CUP The observed runoff values are taken at Thanh My and Nong Son hydrological stations.
Figure 3.13: Comparison between measurement and simulation in Nong Son
APPLICATION OF LINEAR PROGRAMMING
Mathematical-based equations are the cornerstone of the optimization planning process, driving advancements in various solution approaches Key methods include Linear Programming (LP), Dynamic Programming (DP), and Nonlinear Planning (NLP), as well as Mixed Optimization Technique Additionally, other notable approaches encompass Multi-step Approaches, Decomposition and Hierarchical Approaches, Multi-objective analyzes, Decision Support System, Artificial Neural Network Application, Fuzzy Logic Application, and Genetic Algorithm, reflecting the evolving landscape of optimization problem-solving.
Linear programming is a mathematical science that optimizes problems by maximizing or minimizing an objective function, subject to a set of constraints, which are typically expressed as linear equations or inequalities.
With constraints gj (X)≤ bj with j = 1, 2, 3… n
Or we can write the function F(X) in another expression as follow:
F(x1, x2… xi…xn) obtains minimum or maximum value With constraints: g1 (x1, x2,…, xi,…, xn) ≤ b1 g2 (x1, x2,…, xi,…, xn) ≤ b2
……… gm (x1, x2,…, xi,…, xn) ≤ bm With variables of the function is vector X = (x1,x2, …, xn) The optimal solutions of an optimal problem is vector: X* = (x1*, x2*, …, xn*)
In the VGTB case study, the optimal variables are identified as the volume of water allocated to various sectors, which play a crucial role in meeting water use demand in the VGTB basin.
- Volume of water allocated for agricultural production;
- Volume of water allocated for domestic use;
- Volume of water allocated for industrial production;
- Volume of water allocated for livestock.
The objective function of the VGTB case study is centered around optimizing net-benefit values derived from water supply to various water users, with the primary goal of maximizing overall benefits This function is specifically designed to evaluate the effectiveness of water supply systems in meeting the needs of different user groups By focusing on net-benefit optimization, the VGTB case study aims to strike a balance between the benefits and costs associated with water supply, ultimately leading to more efficient and sustainable water management practices.
NAP - Net benefits of water supplied to agricultural production;
NDU - Net benefits of water supplied to domestic use;
NIP - Net benefits of water supplied to industrial production;
NLS - Net benefits of water supplied to livestock.
The objective function is quantified for all the scenarios and is built linearly.
For the optimal problem, a system of constraints plays a very essential role to ensure the rationalization of solutions:
- Volume of water supplied to sectors must be lower or equal to demand;
- Volume of water supplied to sectors must be higher or equal demand regulated by the local administrative agencies;
According to predictions by Da Nang and Quang Nam Province, the urban population of the VGTB basin is expected to reach 852,890 people by 2020 Notably, the majority of this population, approximately 88.7%, resides in cities, totaling 756,604 people, while the remaining 11.3% comprises 96,286 people living in towns.
In 2025 the population keeps increasing to 862,442 people with similar distribution rate in municipality and town.
Table 4.1: Population of the urban area in 2020
According to the water supply standard for the population of the municipality and town issued by the Ministry of Construction, the total volume of water required to meet daily domestic demand is approximately 60.24 million cubic meters per year This breaks down into 55.23 million cubic meters per year for the municipality, accounting for 91.7% of the total, and 5.01 million cubic meters per year for the town, making up the remaining 8.3%.
Table 4.2: Water demand in municipality and town in 2020
The rural population is predicted to increase reaching 885,420 people in 2020 and 895,337 people in 2025, which take 50.94% the total population of the VGTB basin.
Table 4.3: Population of the rural area
According to the water supply standard for rural areas and population using water in the basin, the estimated quantity of water required for rural domestic use was approximately 29.09 million cubic meters in 2020 By 2025, this demand is projected to increase to 29.42 million cubic meters, highlighting the need for efficient water management in these regions.
Table 4.4: Water supplied to rural domestic use
The water supply mechanism in Vietnam is governed by the standard TCXDVC 33:1996, issued by the Ministry of Construction, which outlines regulations for supplying water for domestic use and industrial production, ensuring a standardized framework for water distribution and management.
According to the standard of 100 liters per capita per day in rural residences, the total volume of water required for domestic use is estimated to be 89.33 million cubic meters per year in 2020 and 90.33 million cubic meters per year in 2025 Notably, the sub-basin of Tuy Loan River has the highest water demand, accounting for 44.32% of the total, with 39.59 million cubic meters per year in 2020 and 40.03 million cubic meters per year in 2025.
In the contrary, the Ly Ly River basin requires the least volume of water with 4.81 mil m 3 /year in 2020 and 4.86 mil m 3 /year in 2025, equivalent to 5.38%.
Table 4.5: Water demand for domestic use in the VGTB river basin in 2020
Sub-basin Water demand (mil m 3 ) Percentage (%)
The calculation results show that the total quantity of water required to provide the daily needs is 89.33 million m 3 /year in 2020.
The VGTB river basin's crop schedule and corresponding irrigation coefficients have been determined based on the irrigation reports of Quang Nam and Da Nang City, with a focus on the 85% probability frequency, providing valuable insights for effective water management in the region.
Table 4.6: Crop schedule of crops in the VGTB basin
Type of Crop Crop Schedule
Winter-Spring Rice From 20/12 to 25/04
Type of Crop Crop Schedule
Summer-Fall Rice From 25/05 to 15/09
Winter-Spring Maize From 25/12 to 31/03
Summer-Fall Rice From 25/04 to 31/07
In the VGTB, the total area dedicated to crop production stood at 89,363 hectares in 2020 This was comprised of 30,421 hectares for winter-spring rice and 33,982 hectares for summer-fall rice, while maize production occupied 5,164 hectares in the winter-spring season and 7,001 hectares in the summer-fall season Additionally, 12,795 hectares were allocated for sugar cane cultivation, with further breakdowns available for specific sub-basins.
Table 4.7 : Water use criteria of crops
W-S Rice S-F Rice W-S Maize W-S Maize Sugar Cane q M q M q M q M q M
Ly Ly river 2.03 7,098 2.33 8,241 1.77 2,297 1.76 2,429 1.09 5,208 Tuy Loan river 2.27 7,698 2.34 9,846 1.63 2,469 1.75 3,972 1.17 6,439 VGTB downstream 2.28 7,747 2.37 11,275 1.77 2,569 1.88 4,252 1.19 7,243
Table 4.8: Area of crop in the VGTB basin in 2020
Sub-basin W-S Rice S-F Rice W-S Maize W-S Maize Sugar Cane
From the above input data, the volume of water required by agricultural production can be calculated with below results:
Table 4.9: Volume of water supplied to agricultural production in 2020
The criteria of water volume supplied for livestock in the VGTB basin in 2020 are as follow:
According to projections, the VGTB basin is expected to see a significant increase in its livestock population, reaching approximately 4,266,341 cattle and avian by 2020 Notably, this number is anticipated to remain stable, with no expected changes in the total livestock count by 2025, maintaining the basin's livestock population at a consistent level.
Table 4.10: Quantity of cattle and avian in the VGTB basin in 2020
Sub-basin Buffalo Cow Pig Avian
The total water volume required to supply livestock in the Vu Gia - Thu Bon (VGTB) basin is approximately 14.18 million cubic meters per year Breaking down this requirement by sub-basin, the Vu Gia upstream necessitates around 20.7 million cubic meters annually, while the Thu Bon upstream requires about 3.35 million cubic meters The Ly Ly river basin and Tuy Loan river basin demand 2.55 million cubic meters and 0.67 million cubic meters, respectively Meanwhile, the VGTB downstream area requires around 5.53 million cubic meters of water per year to meet its livestock needs.
According to the Ministry of Construction's standard TCXDVC 33-2996, the water supply requirements for domestic use and industrial production vary by sector For brewing, food, and paper production, a minimum of 45 cubic meters per hectare per day is required, while other industries necessitate a lower volume of 22 cubic meters per hectare per day.
Located primarily in the downstream area of the VGTB basin, a total of 10 industrial zones cover an expansive area of 661.59 hectares, comprising Nong Son, Dong Que Son, Dong Thang Binh, DVTS Da Nang, Da Nang, Dai Hiep, Tay An, Dien Nam-Dien Ngoc, Trang Nhat, and Hoa Cam industrial zones.