This study used the Long-range Energy Alternatives Planning (LEAP) model to conduct an integrated energy system analysis of Bhutan through scenario modelling. LEAP is an energy system model developed by the Stockholm Environment Institute (SEI), which is flexible, user-friendly and free for emerging economies [50]. The objective function of LEAP is cost minimisation from a societal perspective. LEAP can support combinations of top-down and bottom-up approaches to energy system modelling [51], and it is capable of modelling issues beyond technological choices and thus is useful for capacity building applications [52] and to assess Sustainable Development Goal issues [51, 53]. The LEAP model is increasingly being used for low emission development (LED) studies, notably in emerging economies [51, 52, 54-60]. In building the LEAP Bhutan model, its optimisation capability was used for electricity generation expansion alongside its simulation features. The simulation model gives more flexibility to the user to incorporate practical issues.
Scope and limitations
In this study, discussions and analysis are limited to those topics which are deemed pertinent to understanding the overall energy mix, the associated cost and GHG emissions purely from a CN perspective. For this reason, the trend and implications of local air pollutants, are not analysed as the focus is primarily on GHG emissions represented by carbon dioxide equivalent (CO2e) consisting of carbon dioxide (CO2), methane (CH4) and Nitrous oxide (N2O). The study also does not incorporate embodied carbon emissions, or costs associated with infrastructure, houses and buildings including retrofits, which can reduce energy demand. The easiest step for emission reduction from a modelling exercise would be demand reduction through reduction in energy services such as cooking, lighting or passenger-km travelled. But such heavy dependence on absolute end-use demand reduction may not be an appropriate policy tool for a poor emerging country wanting to grow its economy. Notwithstanding this, options for reducing final energy demand (without necessarily reducing end-use demand – a matter of people’s lifestyle) are explored in the LEAP-Bhutan model through energy efficiency, technological shift and fuel switching. Learning rates[1] are assumed only for disruptive, rapidly expanding technologies such as wind, solar photovoltaic and electric transport systems (see Appendix).
It should also be acknowledged that LEAP do not provide the feedback loop of price adjustment arising from a specific mitigation measure [52]. Furthermore, this study also does not analyse the implications of changes to forest cover and its sink capacity is taken as reported by the National Environment Commission [61]. The Constitution of Bhutan has mandated that the State must ensure a minimum of 60% forest cover for all time and until now, forest cover in Bhutan is being maintained at 71% [62].
This study uses the baseline data sets openly accessible in Yangka and Newman [37], which were collected from various sources for the same research project. However, the scope, scenario formulations, and the issues analysed and discussed were distinct from the earlier work. For example, this paper formulated seven distinct scenarios against four in the earlier work. In any case, the scope of modelling an energy-economy system is extensive where scenario formulation comprises different combinations of macroeconomic variables alongside techno-economic variables under an overarching storyline (see the sections below).
Structure of LEAP-Bhutan model and general assumptions
The LEAP-Bhutan model comprised of four main branches: Key Assumptions, Demand, Transformation and Resources, which are further subdivided as per the research requirement and data availability as shown in the Appendix (A1). Key Assumptions consist of the macro-economic parameters which are deployed as drivers of future energy consumption and they are linked at the activity levels under each of the demand sectors. The Demand branch consists of four major demand sectors: Residential, Commercial/Services, Industry and Transport. These are further disaggregated into sub-sectors and end-use type purely based on data availability. Energy consumption in the agriculture sector in Bhutan is negligible compared to other sectors, hence it is accounted under ‘others’ in the commercial sector. The transformation branch is comprised of energy conversions such as electricity generation, coal mining among others. The Resources branch includes primary energy sources and secondary energy sources whether imported or indigenous.
The planning period extends from 2014 to 2050. The emission factors and the global warming potential (GWP) were taken from the technology and emission database (TED) of LEAP. For details on the general assumptions adopted in this study see Appendix (A2). Further information on the data sets and their sources are provided in Appendix (A3)
Distinct features of LEAP-Bhutan model
LEAP-Bhutan model is relatively simple in that there are no fossil fuel extraction and conversion processes such as oil refineries except for coal mining, biogas production, briquette making and the hydropower system. On the demand side, there is no rail system or water ways for transport and limited domestic air ways. Petroleum products are consumed in all the demand sectors; they are imported from India as Bhutan has no oil reserves or oil refineries.
It was noted that there are distinct energy usage patterns in the urban and rural residential areas of Bhutan, which may call for a unique policy intervention, however such disaggregated data are not available in the Energy Data Directory [33]. Further, in the Energy Data Directory, the Industry sector was categorised based on electrical voltage levels such as high, medium, low voltage irrespective of the production system. In this study, the Industry sector was categorised based on industrial output, which was deemed more appropriate for assessing technological options for CN development. Also, in the Industry sector, this study considers charcoal, woodchips and bamboo chips that are being used as reducing agents and as raw materials in the production process, which were not provided in the Energy Data Directory. These goods are mostly imported from India and the required data are synthesised from Department of Industry [63] and Revenue and Customs [64]. All these features lead to distinct parameterization of LEAP-Bhutan model.
Energy Resource and supply
The LEAP model requires the reserve levels for exhaustible resources and yield levels for renewable resources. Such data were calculated from the Bhutan Energy Data Directories [33, 35]. Techno-economic parameters for existing and candidate power plants were obtained from various data sources (see Appendix (A2) and (A3)). The generation profile of the hydropower plants were constrained by the availability curve based on the actual monthly electricity generation obtained from the Druk Green Power Corporation [66]. See Appendix (A2) for details. Increasing exports of renewable energy would theoretically increase the sink capacity for Bhutan’s emissions though this is not considered in this analysis due to the need to also include imports of embedded carbon in a range of products. However, as a general climate policy, exporting renewables to India would be of value to both the global carbon agenda and the local economic growth agenda, while increasing imports from the Indian grid is harmful to both.
Projection of energy demand
Population, Household, GDP, per capita GDP and Sectoral Value Added were used as the drivers of energy consumption that ultimately effects CO2e emissions, the ‘Impact’ in this study under the ImPACT formulation popularly called the Kaya Identity [67-69] for greenhouse gas emissions and ‘T’ being represented by various ‘technologies’ in the LEAP model. These drivers were also used for modelling the low carbon scenario for India [70] and carbon neutral transport system in Iceland [71]. See Appendix (A4) for details. Considering the lack of studies specific to Bhutan to establish an elasticity value between the macroeconomic parameters and the sectoral energy demand, this study assumes an elasticity value of one between the chosen driver and the sectoral energy demand. Yophy, et al. [72] had used GDP elasticity of energy demand as one for the Taiwan LEAP model. It was recognised that elasticity values change over time and are brought about by structural changes and energy efficiency gains, which are separately modelled through fuel substitution and efficiency improvements in the LEAP-Bhutan model over the planning period.
Projection of energy prices
Price of domestic biomass energy was assumed to increase at 3% per annum [39] and for bamboo chips and wood charcoal, which are mostly imported from India for use in the Industry sector were assumed to rise at 4.1% per annum [73]. The prices for petroleum products (see Appendix (A4), table A.13) were projected to follow the international oil price and thus indexed to the price changes calculated from US-DoE [69]. It is reasonable to project the price of petroleum products in Bhutan along with the international oil price projection, since the oil import dependency of India is expected to increase from 74% in 2013 to 91% by 2040 as per the world energy outlook special report [74]. In that report prices of oil, coal and natural gas in India were projected by linking to international prices.
Scenarios storyline and carbon neutral measures
Considering Bhutan’s rising carbon emissions [6, 32, 61] and the need to keep these emissions within the carbon neutral budget of Bhutan, plausible energy-economy pathways were explored. Raskin, et al. [46] note the benefits of scenario analyses stating that ‘Scenarios enlarge the canvass for reflection to include a holistic perspective over space, issues and time’ (pg. 3). Thus, numerous scenarios can be imagined and formulated to explore the future. This study deviates from the usual scenario analyses of comparing alternative scenarios to a baseline in that two groups of scenarios were formulated – Group A and Group B - to provide a clear understanding of the underlying assumptions that drive each scenario and also to expand the scenario space (see following sections). Group A has four different baseline scenarios of growth against the use of a single baseline in the existing energy-economy planning literature, but they are all plausible baselines. Group B has the corresponding CN scenarios; the aim here is to investigate what it will take for Bhutan to sustain its CN pledge if it pursues those plausible growth pathways defined in Group A.
The scenarios under group A emerge from a macroeconomic outlook and entail a distinct energy system pattern – in terms of type and amount of energy consumed, type of primary resources extracted, and type of demand technology used by the demand sectors. Group A comprise the Business-as-usual (BAU) scenario, the high economic growth (HGDP) scenario, the low economic growth (LGDP) scenario and the Nationally Determined Contributions (NDC) Plan scenario. The NDCPlan scenario attempts to depict some of the key aspects of the NDC committed to by Bhutan, which had outlined nine broad strategies [6] but without specific targets. This present study assigned some reasonable quantitative targets for modeling purposes based on past trends and other national level documents, with an intention to reduce carbon emissions.
Group B consists of the CN counterparts for each of the scenarios in Group A though with a distinct sector in focus (e.g. the ‘advanced technology’ scenario focusses just on the transport and industry sectors and the ‘modern fuel’ scenario focusses just on the residential and commercial sectors). Group B is intended to decouple the growing Bhutanese economy and its energy demand from GHG emissions, thereby holding the emission levels within the sink capacity. Decoupling is explained in Newman [14]. To formulate Group B scenarios, distinct and exclusive CN measures are specified as shown in Table 1. Group B scenarios are therefore more challenging to the economy than Group A but are not seen as highly radical or beyond possibilities.
The various scenarios will be examined to see how they keep within the sink capacity of 6.3 MT (million tons) of CO2e – this is the total forest sink capacity that can neutralize any carbon emissions produced in Bhutan – apparently taking a territorial approach. Combinations of mitigation measures will also be done to form Group B scenarios.
Group A scenarios
Group A scenarios consists of the BAU, HGDP, LGDP and the NDCPlan scenarios. The BAU scenario represents the 2014 energy-economy trajectory and assumes a GDP growth rate similar to that witnessed in the past two decades (refer table 2). Major policy interventions are not anticipated except for the general trends such as the declining usage of fuelwood in the residential and service sectors due to achievement of nation-wide electrification and some push in the public transport sector. The HGDP scenario represents the high economic growth rate of 10% per annum until 2025 and thereafter sustaining at 7.8%. In recent years Bhutan has achieved such high growth as a result of the commissioning of mega-hydropower projects. The LGDP scenario represents low economic growth of 5.6% which further declines to 2.5% by 2050. Bhutan has also witnessed such low economic growth in 2012.
Group B scenarios and mitigation measures
For each of the scenarios under Group A, a corresponding CN scenario was formulated based on the selection of mitigation measures outlined below.
This study specified four aggregated measures based on data availability, suitability to Bhutan’s context and the emerging global vision shown in Table 1 with detail descriptions provided in Appendix (A5.3).
Selection of the mitigation measures
The mitigation measures outlined in Table 1 above were then deployed one by one onto the BAU scenario to obtain the cost of mitigating a tonne of GHG through the cost-benefit summary report in the LEAP model. It was observed that none of the measures except for the F8050 were effective enough on their own to keep the GHG emissions within the sink capacity, despite some of the measures being no-regret options (i.e. options with no cost or negative cost, meaning they save money). Given these limits, the mitigation measures were combined to form groups of cost-effective CN measures to hold the emission level within the sink capacity at the lowest cost of mitigation during the planning period. The process of moving from one combination to the next is based on the cost of mitigation shown in Figure 1. The cheaper options (e.g. 1st and 2nd lowest) were combined first to see if the emissions level remains within the sink capacity, if not, then the 1st is combined with the 3rd lowest and so on. If the combination of any two measures failed to keep the emission levels within the sink capacity, then the combination of three measures were applied onto the scenario under investigation. This leads to the formation of Group B scenarios.
Group B – Carbon Neutral Scenarios
The CNBAU represents the Carbon Neutral BAU scenario, which includes the two mitigation measures ‘IS + ATECH’ that form the least cost option in holding the emissions level within the sink capacity. Similarly, CN_NDCPlan represents the Carbon Neutral NDCPlan scenario with ‘IS + MF30’ as the combined measures. The CNHGDP represents the Carbon Neutral high economic growth (high GDP) scenario with ‘IS + MF30 + F8050’ as the combined measures. Under the LGDP scenario, the emissions level remains well below the sink capacity, thus application of the mitigation measures does not arise. Low growth may not be an effective policy, however, as economic and social goals are largely pushing the country towards a high growth future based on the need to create opportunities and growth provides these. Growth is therefore driving the politics inevitably in this direction as in most emerging countries looking to break out of poverty. Its implications for energy need to be addressed.
[1] Learning rates refer to a percentile value which implies reduction in the cost of technologies when its production amount doubles