Inter-Basin Water Transfer Projects and Climate Change: The Role of Allocation Protocols in Economic Efficiency of the Project. Case Study: Dez to Qomrood Inter-Basin Water Transmission Project (Iran) ()
1. Climate Change and Long Term Variation of Rivers Flow: A Forbidden Factor
Owing to the rising tide of world population and living standards, we can claim that actually a regime of water shortage has been established in all around the world. It’s obvious that this regime is more severe in some areas. Nowadays, the growing water demand has resulted in evaluation of even costly solutions and applying them. As an example, Water resources Managers attempt to provide water in developing areas with water transmission from a rich basin to a poor one. However, the vague aspect of these projects is the question that “Do the current transmission patterns optimize water allocations in long term and consider all stakeholders’ benefits living in both up and downstream of a water transfer project? Are these protocols thoroughly reliable to face long term effects of Climate Change?” To answer these questions we should consider these two issues:
1) Long effects of Climate changes on water resources.
2) Water transmission protocols and their performances.
Climate prediction, as a new science, faces some difficulties caused by uncertainties of the natural system. For decades, predictions are done based on greenhouse gases estimation. It goes without saying that greenhouse emission is a complicated socio-economic issue with many ambiguous effective factors on the emission rate. On the other hand, the results of different GCMs are considerably not the same for a specific area. Due to these reasons, if a research is supposed to be helpful in water resource management, a range of climate scenarios and GCMs should be constructed to capture a desirable part of the uncertainty space. Many researchers have done immense researches by using multi model projection such as Van Oldenborgh et al., Chikamoto et al., Kunstmann et al., Serrat-Capdevila et al., Andersson et al., and etc. [1-5].
From another point of view, an overwhelming majority of water allocation agreements are established based on long term average flow. This is in a case that political issues cast a shadow over these agreements while technical aspects of water engineering are on the margins of them. Moreover it’s important to know that many of water transfer projects had been planned before the announcement of the Climate change Concept. In the twentieth century, 145 international agreements on water use in trans-boundary Rivers were signed; and almost 50% of these agreements cover water allocation issues [6]. Although variability is an important characteristic of river flow, (even with or without considering climate change effect), an overwhelming majority of these water allocation agreements do not take into account the hydrologic variability of the river flow [7]. It’s obvious that these agreements never discuss sustainable development, system optimization, justice, and environmental demand. In this research we are going to present a systematic approach which can help water resources management to select an optimum allocation with a superior performance in the face of climate change.
2. Probability Space and GCM-Scenario Combinations
As it was mentioned in the previous chapter, final result of a rough climate prediction is completely depended on the GCM and the climate scenario which are used, and these results are not the same for different GCM-scenario combinations. Applying several GCMs is a common accepted method in hydro-climatic researches; for example this method was used by Serrat-Capdevila et al. to model climate change impacts on the riparian system hydrology of San Pedro Basin (Arizona/Sonora) [4]. Furthermore Andersson et al. modeled Impacts of climate change on Okavango River (shared by Angola, Namibia and Botswana) by applying these methods [5]. Generating scenarios for exploring a probability space is an old tradition in decision-making in water and energy context [8-11]. In this research we use different climate scenarios to find the effects of Climate change in Dez and Ghomrood (Qomrood) basins. As each Global Circulation Model (GCM) has its individual results for a particular Green House Emission scenario (GHE), we’ve decided to cover the majority of valid GCMs and GHEs by using nineteen GCMs and four GHEs, This idea results in development of a comprehensive space of probable climate condition.
The geographic position of Dez and Ghomrood (Qomrood) basins in Iran is illustrated in Figure 1. Dez basin is located in the western Zagros massif by high precipitation and significant seasonal rainfall; on the other hand, Ghomrood (Qomrood) basin is located in the central arid region. An under construction transmission link will collect water from five local branches in higher altitude and will transfer flow to Ghomrood river where two reservoir (Kuchrey and Golpaygan dams) regulate and dispense the flow. A diagram of the system is shown in Figure 2. MAGICC-SCENGEN (Model for Assessment of Green-
Figure 1. General geographic location, cells, and microcells which are generated in Dez and Qomrood basins by MAGICCSCENGEN.
Figure 2. Schematic scheme and view of allocation model’s interface window of Dez to Qomrood water transmission project, more information are shown in Table 1.
house-gas Induced Climate Change) and its Nineteen GCMs have been used to calculate the change of precipitation in global scale. The resolution of MAGICCSCENGEN is a mesh of 2.5 × 2.5 degree cells. Figure 1 shows that how forty eight 2.5 × 2.5 degree cells covers Iran and environs, while Dez and Qomrood basins are located in no.19 and no.20 cells. Clearly, the GCMs cannot resolve the spatial structure of climate at the subbasin scale used in the hydrological model. To down scaling precipitation changes in local scale we used reverse-distance factors by utilizing results of neighboring cells given from GCMs in global scale. This method is applied by Andersson et al. to down scale precipitation results of GCMs on Okavango River basin [5]. To forecast monthly-scale precipitation, we have used GCMs results for monthly variation and compared these with historical monthly distribution to approach a basic monthly distribution function for each GCM-GHE scenario. Mitchell et al. have employed this method to forecast Europe and the globe climate factors [12]. In order to make more detailed grid network in the basin area, a minor mesh involving 0.5 × 0.5 degree microcells was generated and was utilized for downscaling.
Forecasted precipitation is used to develop runoff characteristics by utilizing a rainfall-runoff model. In this research historical data of rainfall and flow have been applied to develop monthly flow generator via a linier equilibrium. These generators are used to develop runoff for stochastic series of rainfall. Such simplified models, linier or non linier, are employed in several researches to forecast runoff in similar cases. Gardner employed exponential equilibrium to assess annual runoff in catchments with a wide range of climatic conditions [13]. This method also has been applied by a wide range of researchers like Graham, Chen, Benestad, and Carter [14- 17]. It’s very clear that many factors like land use, agriculture and irrigation patterns which have effects on basin’s hydrologic conductivity, are variable in long time. However, in this level of research scope, these uncertainties are inevitable and we neglect their impacts.
Table 3 illustrated final results for all GCM-Scenario combinations for 2050. In this research, the results of GCM-GHE scenarios have been classified in five groups Including Very Optimistic, Optimistic, Moderate, Pessimistic, and Very Pessimistic forecasting. Each group is weighted by the proportion of its GCM-GHE combinations to the whole probable space (see Table 2). All these GCMs have been used in IPCC Third Assessment Report (TAR). One GCM of each family is selected as a demonstrator of the group behavior. As it is shown in the Table 2, GCMs that we used in research are: GFDLCM-2.0, GISS-EH, UKHAD-GEM, GFDLCM-2.1 and FGOALS- 1G. Final results are Conformable with Third Assessment Report (TAR) generally. More than 68% of GHEGCM scenarios result a reduction between 6 to 28 percent in rainfall and consequently in the same order for runoff (see Table 3). TAR assesses reduction between 10 to 40 percent for 2090-2099 duration, relative to 1980