Analysis of water resource benefits due to power grid interconnections using the virtual water method

1 Introduction

Freshwater stress has become a severe problem globally.It affects human health and economic development on every continent.Approximately 4 billion people experience high water stress for at least one month of the year [1],and it has been predicted that 700 million people worldwide could be displaced by intense water scarcity by 2030 [2].Although a growing number of studies have investigated freshwater vulnerability in various spatiotemporal contexts,most focused on physical freshwater from a territorial perspective,considering water availability,water demand,and water consumption within a city,region,or country [3-7].Thus,these studies did not consider indirect water consumption embedded during trading across geographic boundaries.Not only is the water stress in the consuming areas well beyond the local demand,but it is also related to the total water demand of all production sectors in the producing areas [8].

Energy production uses approximately 52 billion m3 of fresh water every year [9],posing a significant challenge to the energy sector.Power generation exerts very high water demand in the energy sector [10]mainly because all coal-fired power plants across the world are estimated to consume approximately 19 billion m3 of water every year [11].With the increasing population and economic development,the imbalance between electricity demand and water resources utilization is bound to worsen.

The term “virtual water” encompasses the hidden flow or indirect consumption water [12].It refers to water used for the production of goods and services,and can be transferred among different regions through trade [13].Previous literature on virtual water mainly focused on quantifying water used in major trading products,such as food and clothing,and other international services [14-15],but little is known about how water stress changes due to electricity generation and exportation.Many researchers have also analyzed the water-electricity nexus or waterenergy nexus by characterizing the interdependencies of the production and consumption of water and electricity [16-20].The direct method of analyzing this nexus involves estimating how much electricity is consumed for waterbased processes and how much water is consumed for electricity generation,using life-cycle analysis models or technology-based estimation models [21-23].The indirect method of analyzing this nexus involves estimating material and monetary flows by linking the electricity and water systems with the social and economic systems and constructing an input-output analysis or computable general equilibrium models [24-26].The direct method of analyzing the water-electricity nexus is usually a bottomup approach,while the indirect method is typically a topdown approach.With electricity trading across countries and regions becoming more common,a growing number of studies have attempted combining the water-electricity nexus and concept of virtual water to investigate virtual water flows involved in electricity generation and trading[27-30].However,these studies only quantified the current virtual water flow with the current energy mix,mainly with regard to thermoelectric power plants.

Studies of virtual water provide insights into the water resources benefits of power grid interconnections.The following issues,however,still need to be explored:1)changes to virtual water when clean energy dominates the energy mix,2) improvements to water resources distribution on account of virtual water flow through largescale interconnected electricity grids,and 3) supporting information for policy makers with regard to the price mechanism of virtual water for establishing electricity trading systems or electricity-carbon trading systems.In this paper,we propose a framework of water conservation accounting in power systems using the virtual water method.By defining the water scarcity index as a price mechanism indicator,the improvement in water resources allocation through power grid interconnections is evaluated.The remainder of this paper is organized as follows.In section 2,we introduce the water conservation accounting method used in this work,and the detailed mechanism to calculate the water resources benefits resulting from electricity interconnections is presented.In section 3,numerical analysis is conducted for a case study of the Africa-Asia-Europe grid interconnection scenario.Finally,section 4 concludes this paper.

2 Water conservation due to power grid interconnections

2.1 Water conservation accounting in power systems by the virtual water method

Virtual water in the context of this work is defined as the water consumed in the generation of a certain amount of electricity.As shown in Fig.1,a certain amount of electricity eAB is generated in region A and delivered from region A to region B through grids.The virtual water consumed during this power generation can be deduced using the amount of electricity,power generation mix of region A,and average generation water intensity (water consumption per unit of electricity generated).The vector of average generation water intensity of region A can be written as

Fig.1 Virtual water of power transmission between regions A and B

where dA,k (k=1,2,...,K) is the average generation water intensity of the k-th type of power plant (e.g.,gas power plants,wind power plants,hydropower plants,etc.),and K is the total number of power plants.Given the vector of power generation mix of region A,the virtual water of power generation in region A can be written as

where SA represents the vector of power generation mix of the transmitted power from region A to B,expressed as

sABM is the proportion of the power generated by the m-th energy type and transmitted from region A to B,and M is the total number of energy types for power generation.Therefore,we have

Obviously,the virtual water given in (2) is determined by the amount of delivered power and the water consumption per unit of electricity generated.Notably,the amount of virtual water increases as the amount of generated electricity increases.Moreover,the amount of virtual water will be large if water-consuming power generation techniques are used in region A.

To quantitatively analyze water resource conservation by the power transmission between regions A and B,we then calculate the amount of virtual water by assuming that eAB is the amount of power generated in region B.The vector of average generation water intensity and the vector of power generation mix of region B can be respectively expressed as

and the virtual water of power generation in region B can be written as

where sBm is the proportion of power generated by the m-th energy type in region B.Therefore,according to (2) and (6),the water conservation caused by the power transmission VAB can be represented as the difference in the virtual water in regions A and B,such that

It is obvious that the water conservation result presented in (7) can be positive,zero,or negative.If wAB,A≥wAB,B,the water consumed in region A for generating the electricity amount eAB is larger than that of region B.In this case,the power transmission from A to B cannot facilitate water conservation.If wAB,A＜wAB,B,conversely,the water consumed in region A for generating the electricity amount eAB is smaller than that consumed in region B,and thus,water resources are conserved in this case.

2.2 Water conservation in multi-region power transmission

The above-mentioned deductions for water resource conservation can be extended to multi-region cases.As shown in Fig.2,consider a scenario of large-scale grid interconnections,where multiple regions are involved in the power transmission.Some of the nodes serve as either transmitting nodes or receiving nodes,whereas the remaining work in both capacities.Due to the power transmission between regions,the total water conservation is expressed as the cumulative conservation in all branches,such that

where Vij is the water resource conservation achieved due to the power transmission between i and j,and U is a set of regions in the analysis.

Fig.2 Multi-region power transmission

In real cases of power transmission between multiple regions,the power generation mix of the transmitted power from a certain region i may not be the same as the local power generation mix of that region.For example,the electricity delivered from region i to region j is generated from a point power source (e.g.,a hydro,wind,or solar PV plant),but the power in region i for local use is provided by the local grids,which are composed of with various power plants.Therefore,it is unnecessary that Si=Sij.

The power transmission based on grid interconnection can not only reduce the total water consumption of the entire system developed in (8),but also relieve the water resources stress in water-deficient regions.The water resources allocation of a certain region can also be analyzed by the virtual water method.As shown in Fig.3,region B receives power from region A and also transmits power to region C.According to the aforementioned analysis,the water conservation of the entire system can be represented as

From the perspective of region B,water is saved by receiving power from region A,but water is also consumed by generating power for region C.The summation of the saved and consumed water resources in region B can be represented by the second and third terms of (9),such that

Obviously,by considering electricity import and export,region B consumes water if VB ＞ 0 while it saves water if VB ＜ 0.By considering a region with multi-import and multiexport electricity,the result can be extended by a general expression,as follows:

Fig.3 Water saved and consumed in region B

The virtual water reveals the value of water in terms of electricity trading,and could be used in future electricity trading markets.Typically,the price of water is not completely resources- or market-based.By using the water scarcity index as an indicator for the virtual water price,and using the summation of the saved and consumed water in region B given in (11),the total benefit of water resources allocation under power grids in region i can be expressed as

where pi is the water scarcity index of region i.Vi is negative if water is saved in region i,and therefore,Si would also be negative,indicating the positive water allocation effect.Moreover,the absolute value of Si indicates the extent of the water resources allocation effect.

3 Case study

In this paper,the Africa-Asia-Europe power grid interconnection scenario is taken as a special case to calculate the virtual water flow of intercontinental power transmission.The goal here is to evaluate the efficiency of the proposed method for water resource conservation.

Based on the African Energy Interconnection Planning Study [31],Africa’s energy demand will grow rapidly in the future.From 2015 to 2050,Africa’s primary energy demand is predicted to expand from 1.12 billion standard to 2.41 billion standard coal tons,with an average annual growth rate of 2.2% and accounting for 9.3% of the global total.From the point of view of terminal energy,Africa’s electricity requirement will increase from 9.5% to 28%.According to [32],from 2015 to 2050,the electricity generation in Europe and Asia will increase from 4838 TWh to 8360 TWh and from 11316.6 TWh to 24873.7 TWh,respectively.The corresponding average annual growth rate would equal 1.6% and 2.3%.Under the premise of clean development,large-scale clean power supply and intercontinental power transmission will be essential to meet the growing energy demand of all continents in the future.Africa is rich in clean energy resources.The Congo River and Nile River provide sufficient conditions for hydropower development in Central and East Africa,respectively.Considering the gradual development in intercontinental power grid connections in the future,hydropower transmission will be realized in Central and East Africa to provide clean electricity for northern,western,and southern Africa.In addition,North Africa and West Asia have mature solar bases and will likely deliver solar energy to provide clean electricity to Europe and North Africa,respectively.Given East Africa’s abundant water resources,it will also be able to provide intercontinental clean power transmission to West Asia.Based on the above scenarios,the virtual water flows of East Africa,West Africa,Southern Africa,North Africa,Central Africa,West Asia,and Europe were calculated for the two time nodes of 2030 and 2050,according to the electricity mixes of and power transmission among these regions.Using the water scarcity index,the water conservation benefits of virtual water flow under the Africa-Asia-Europe power grid interconnection can be indicated,as explained below.

3.1 Africa-Asia-Europe interconnected power flow and power generation structure

Fig.4 Electricity topology under the Africa-Asia-Europe power grid interconnection

To calculate the virtual water flow of power transmission and evaluate the water-saving effect of grid interconnection,this study assesses the Africa-Asia-Europe cross-continent power flow based on the future grid interconnection construction plan [31-32](Fig.4).By 2030,the intracontinental power grid interconnection in African will form three major power transmission channels,namely East Africa-South Africa,Central Africa-West Africa,and Central Africa-Southern Africa,with an average annual transmission of 16 TWh,48 TWh,and 4 TWh,respectively.At the intercontinental level,the electricity sent by West Asia to North Africa and by North Africa to Europe will equal 4.5 TWh and 16.5 TWh,respectively.By 2050,electricity transmission in these three power transmission channels(East Africa-South Africa,Central Africa-West Africa,and Central Africa-South Africa) will increase to 32 TWh,154 TWh,and 40 TWh,respectively.Moreover,three new power transmission channels (Central Africa-East Africa,Central Africa-North Africa,and East Africa-North Africa) will be constructed,their average respective annual electricity transmission being 54 TWh,60 TWh,and 48 TWh,respectively.At the intercontinental level,the electricity sent from West Asia to North Africa and from North Africa to Europe will increase to 14 TWh and 126 TWh,respectively.Moreover,the newly built East African-West Asian power transmission channel has a delivery capacity of 16 TWh.

Based on the future power development plan [31-32],the power generation structures in Africa,Europe,and West Asia are shown in Table1.Considering the probable situation of power transmission in the future,East Africa and Central Africa will both provide electricity generated from hydroelectric power,and North Africa and West Asia will provide that generated from solar power.

Table1 Power Generation Structures of Africa,Europe,and West Asia

2030 Coal (%) Oil (%) Gas (%) Wind (%) Solar (%) Hydro (%) Bioenergy (%) Nuclear (%)East Africa 0 8.2 15.2 1.68 11.79 41.64 19.52 0 West Africa 0 19.4 25.0 0.46 13.25 31.22 10.66 0 South Africa 48.2 1.8 8.3 7.33 13.4 8.46 2.76 0 North Africa 3.9 15.2 53.9 2.26 20.11 2.99 1.72 0 Central Africa 0 3.1 3.8 0.1 0.44 90.19 2.39 0 West Asia 3.56 6.83 28.63 3.34 45.65 3.93 0.62 7.45 Europe 1.27 0 18.18 31.53 13.7 15.91 6.92 12.49 2050 Coal Oil Gas Wind Solar Hydro Bioenergy Nuclear East Africa 0 1.4 14.2 11.73 25.53 31.18 14.68 0 West Africa 0 5.7 24.1 0.4 39.48 20.47 9.84 0 South Africa 20.2 1.2 8.9 14.75 22.58 8.57 1.12 0 North Africa 2.1 11.4 30.1 4.55 48.86 1.75 1.27 0 Central Africa 0 0.2 2.1 0.03 0.11 95.68 1.82 0 West Asia 1.57 5.6 18.43 3.87 61.02 3.25 0.59 5.66 Europe 0 0 6.61 47.17 16.01 16.17 6.61 7.44

Fig.5 Virtual power flow topology under the Africa-Asia-Europe power grid interconnection

3.2 Virtual water flow and allocation analysis under the Africa-Asia-Europe interconnection power grid

Based on the future Africa-Asia-Europe interconnection construction plan [31-32],the power transmission capacity,power generation structure,and virtual water flow corresponding to the probable electricity generation in this scenario can be calculated.In this example,the unit water consumption factor (m3/MWh) of coal,oil,and gas is set to 2.65,3.2,and 1,respectively,and the unit water consumption factor of wind,solar energy,water,biomass and nuclear energy is 0,1.6,0,1.8,and 3,respectively [10].

As shown in Fig.5,by 2030,the three major transmission channels (East Africa-South Africa,Central Africa-West Africa,and Central Africa-Southern Africa)will produce a large amount of virtual water flow (Table2),the respective absolute annual amounts being 24.32 million m3,53.58 million m3,and 6.08 million m3.At the intercontinental level,the annual virtual water flows from West Asia to North Africa and from North Africa to Europe are 3.77 million m3 and 1.2 million m3,respectively.By 2050,annual virtual water flows in the East Africa-South Africa,Central Africa-West Africa,and Central Africa-South Africa channels will increase to 24.77 million m3,116.71 million m3,and 30.96 million m3,respectively.The annual power transmitted from Central Africa to East Africa,Central Africa to North Africa,and East Africa to North Africa is 29.93 million m3,56.3 million m3,and 45.04 million m3.At the intercontinental level,the annual amount of electricity transmitted from West Asia to North Africa and from North Africa to Europe increased to 11.72 million m3 and 9.1 million m3,respectively,and the corresponding value from East Africa to West Asia is 13.27 million m3.

Table2 Virtual water flow under the Africa-Asia-Europe power grid interconnection

Absolute value of virtual water(million m3/yr )2030 2050 East Africa-South Africa 24.32 24.77 Central Africa-West Africa 53.58 116.71 Central Africa-South Africa 6.08 30.96 Central Africa-East Africa 0 29.93 Central Africa-North Africa 0 56.3 East Africa-North Africa 0 45.04 West Asia-North Africa 3.77 11.72 North Africa-Europe 1.2 9.1 East Africa-West Asia 0 13.27 Total 88.95 337.80

Fig.6 Total virtual water flow based on the electricity trade in Africa-Asia-Europe

The estimated water saving effect of the Africa-Asia-Europe power transmission (i.e.,the absolute amount of total virtual water flow) is shown in Fig.6.By 2030,the total annual virtual water flow due to power transmission will reach 88.95 million m3.By 2050,the estimated annual water saving effect will reach 337.8 million m3,translating into an increase of 279% over the estimated amount for 2030.

This study used the water scarcity index to assess the water conservation benefits of virtual water flow generated by the Africa-Asia-Europe grid interconnections.The standardized data for water stress,that is,freshwater withdrawal as a proportion of available freshwater [33],used to calculate the water scarcity index are shown in Table3.The Africa-Asia-Europe grid promotes clean power transmission through grid interconnections.According to the virtual water flows calculated above,annual water saving benefits of 2193.68 million m3 and 44653.24 million m3 can be achieved by 2030 and 2050,respectively(Table4).Intracontinental and intercontinental power grid interconnections will greatly promote the development of renewable energy in Africa and West Asia.The watersaving benefits of clean electricity transmission not only provide strong support for water resources protection,but also promote low-carbon transformation and sustainable development in Africa and West Asia.

4 Conclusion

The global energy system is undergoing transition.In the future energy system,the energy production pattern will gravitate from fossil fuel energy to clean energy.The energy distribution pattern would transit from local balancing to cross-country,cross-continental,and global allocation.Moreover,final energy consumption would be electricitycentric.Therefore,it is necessary to study the linkage between electricity and water to understand the impacts on clean energy production and to evaluate the water resource benefits based on electricity interconnections.This study expands the traditional virtual water concept to define the generalized virtual water flows of a power system with positive,zero,or negative values.Thus,this study proposes a generalized virtual water tool to estimate the water conservation benefits of clean power transmission under grid interconnections from both the production and the consumption sides.

Taking the future Africa-Asia-Europe interconnection scenario as an example,the water conservation benefits ofclean power transmission are calculated.Clean electricity transmission in six sub-regions of Africa,Europe,and West Asia can allow an estimated annual total water savings of 88.95 million m3 by 2030 and 337.8 million m3 by 2050.The Africa-Asia-Europe grid interconnection can promote the development of renewable energy in Africa,while achieving greater water conservation benefits.

Table3 Water scarcity index of different regions

Region Water scarcity index South Africa 10.85 West Africa 3.91 East Africa 9.73 North Africa 259.43 Europe 563.53 West Asia 666.32

Table4 Conservation benefits of virtual water flow under Africa-Asia-Europe power grid interconnections

Conservation benefits of virtual water flow (million m3/yr)2030 2050 East Africa-South Africa 264 269 Central Africa-West Africa 210 457 Central Africa-South Africa 66 336 Central Africa-East Africa 0 291 Central Africa-North Africa 0 14606 East Africa-North Africa 0 11685 West Asia-North Africa 978 3040 North Africa-Europe 676 5128 East Africa-West Asia 0 8842 Total 2194 44654

This paper focuses on the virtual water flow and water resources benefits accruable from large-scale development of clean energy and electricity interconnections across countries and continents under the long-term for specific scenarios.The results of this study also underscore the possibilities of water resources allocation through power grid interconnections.However,this study did not consider certain realistic factors such as the various local and external power supply structures involved in short-term power system operations,and the conditions of transmitting or receiving power during peaking and valley periods.In addition,it is important to assess the possibility of integrating water,electricity,and even carbon prices under a systematic trading market,and to identify the coordinating mechanisms that would support such trading markets.These topics will be investigated in a future study.