Introduction

Air quality management and climate change mitigation are two inexorably linked environmental challenges of the twenty-first century. Addressing them in a coordinated manner can simultaneously slow down the rate of climate change and protect human health and ecosystems, including agriculture. Yet, air pollutants and greenhouse gases (GHGs) and their impacts are often considered independently in both scientific and policy spheres. CO2 is widely recognized as the primary driver of global warming and climate change. However, studies have shown that short-lived climate pollutants (SLCPs)—including methane (CH4), black carbon (BC), and tropospheric ozone (O3)—contribute to near-term climate change, as well as adverse impacts on human health and agriculture (IPCC 2014; Ramanathan and Carmichael 2008; UNEP/WMO 2011). Growing evidence suggests that to reduce global warming and remain under a target of 2 °C rise, it is essential to take a coordinated action, without any delay, for reductions in both CO2 and SLCPs concurrently (Bowerman et al. 2013; Hu et al. 2013; Rogelja et al. 2014; Shoemaker et al. 2013; UNEP/WMO 2011). To help address this, several governmental and non-governmental organizations (including the US Environmental Protection Agency, the World Bank, and the Arctic Council) are taking action on SLCP mitigation (Pierrehumbert 2014). In addition, a separate United Nations entity, the Climate and Clean Air Coalition (CCAC), was formed in 2012 as the only global voluntary organization dedicated to promoting and implementing early mitigation of SLCPs by supporting their integration into existing national planning (CCAC 2016b). Clearly, an integrated approach to addressing air quality and climate change as part of the policy process will offer a great opportunity to contribute to meeting sustainable development goals (SDGs), although they are not explicitly stated as such (Haines et al. 2017).

Several studies have identified Asia as the single largest source of global BC emissions from contained combustion, accounting for more than half of all such emissions (Ramanathan and Carmichael 2008; UNEP 2011; USAID 2010). The Hindu Kush Himalayan (HKH) region in South Asia is especially vulnerable to the impacts of SLCPs, in particular to BC. This region is the source of ten large Asian river systems which provide water, ecosystem services, and the basis of livelihoods to more than 210 million people in the mountains and 1.3 billion people downstream (Beniston, cited in Rasul 2014). Studies have shown that glaciers in the HKH region have been retreating and will continue to melt at higher rates if the increase in emissions continues (Ramanathan et al. 2008; Rose 2012; Yongjian et al. 2015). While some of the atmospheric changes in the HKH region are driven by the global increase in atmospheric GHG concentrations, approximately 50% of the warming on the Himalayan–Tibetan plateau has also been attributed to BC (Menon et al. 2010; Ramanathan et al. 2007; UNEP/WMO 2011). This glacial melting in the HKH region is due not only to a temperature increase in CO2 but also to the aerosols that arise mainly from burning biomass and fossil fuel combustion (Gustafsson et al. 2009; Ramanathan et al. 2007; Rose 2012; Sadavarte et al. 2016).

Nepal, located between two of the world’s biggest BC emitters, China and India (Ramanathan and Carmichael 2008), is particularly vulnerable to the impacts. Thus, domestic action will be insufficient and regional cooperation will be needed to reduce the impacts of SLCPs. However, there is still a data gap on SLCP emissions in Nepal based on their activity levels, emission share, temporal and spatial variation, and quantification of their impacts. Existing policies and plans are largely designed to indirectly support or generate co-benefits for air pollution mitigation without an explicit SLCP-focused policy and planned interventions (Gyawali 2016). Past studies of SLCPs in Nepal are usually either concerned with transboundary atmospheric brown clouds (Lüthi et al. 2015; Rose 2012), or focused at city level and reflect localized data as in the following papers: Rupakheti et al. (2016); Kim et al. (2015); Putero et al. (2015); World Bank (2014); Shrestha et al. (2013b); ICIMOD (2012); Dhimal et al. (2009); CEN/ENPHO (2003). The World Bank reported that the mean annual ambient PM2.5 concentration in Nepal was 46.09 μg/m3 in 2013, and the PM2.5 concentration in Kathmandu in 2013 was 49 μg/m3 (WB/IHME 2016; WHO 2016). A more recent study shows that the daily mean PM2.5 and BC concentrations in Kathmandu valley, due to the transport sector, can reach 124.76 μg/m3 and 16.74 μg/m3, respectively, during spring (Shakya et al. 2016). Other studies suggest that the urban centers are most vulnerable to impacts of air pollution with pollution levels substantially above WHO guidelines (CANN 2014; CES 2016; DoE 2016; Gautam 2010; ICIMOD 2012; WHO 2016; World Bank 2014). All these studies call for the control of air pollution to reduce its adverse impacts. For this to happen, a proper understanding of the impacts of SLCPs mitigation measures is paramount for evidence-based policy decision-making. In this context, it is very important for Nepal to undertake an assessment of the potential measures that could be taken to mitigate the impacts of SLCPs.

The objective of this study is to provide a guideline or benchmark for the formulation of a national action plan to be undertaken by each of the respective stakeholders and practitioners. Application of the scenario-based approach using the Long-range Energy Alternatives Planning System–Integrated Benefits Calculator (LEAP-IBC) analytical tool for evaluation of the mitigation of SLCPs in this study also provides an example for devising mitigation strategies in other countries in the HKH region.

The paper is organized as follows: the “Methodology and modeling framework” section briefly discusses the methodological framework dealing with mitigation scenario, impact assessment method, and data sources. The “Results and discussion” section presents scenario results and discussion, followed by our conclusion in the “Conclusions and policy implications” section. The detailed methodological framework is provided in the Supplementary Materials for this paper.

Methodology and modeling framework

This study has been carried out in close compliance with the national SLCP planning process as given by CCAC in the SLCP National Planning Guidance Document (CCAC 2016a). The initial phase comprised a rigorous literature review and consultations with various governmental as well as non-governmental stakeholders and experts on air pollution. This phase contributed to assessing the current situation regarding emissions and inventory development. It also provided insights on current activities, policies, plans, and institutional frameworks related to SLCP mitigation. Second, emission estimates were derived for the Reference (REF) scenario, beginning in 2010 and extending to 2050 using a bottom-up approach. Third, an analysis of mitigation options suitable in the context of Nepal was carried out from various documents, with priority given to those identified by a UNEP/WMO assessment report and reports on sectoral mitigation options for SLCPs in the HKH region (MOPE 2014; Sharma 2014; UNEP/WMO 2011; USAID 2010; USEPA 2012). A baseline inventory of emissions was developed. Then, the scenarios were analyzed using the LEAP-IBC modeling framework, with economic and demographic parameters taken as drivers of anthropogenic activities and emissions. Finally, the results were ratified by governmental and non-governmental stakeholders and experts in a validation workshop.

Emission mitigation scenarios

Based on the economic and demographic situations as primary driving factors, SLCP emission projections and their impacts were examined under Reference (REF) and Policy (POL) scenarios. The year 2015 was taken as the base year for results analysis. Agricultural, commercial, and industrial activities were assumed to be dependent on respective gross value added (GVA) in each emitting sector, while the residential sector and waste outputs were assumed to be dependent on population. The transport sector, on the other hand, is dependent on both economic and demographic parameters for freight and passenger transportation, respectively. The economic and demographic data were retrieved from CBS (2012, 2014), NPC (2017), and World Bank (2013). The assumption of the REF scenario is that the future trend will follow the same path as the current one with no change in current policies. The POL scenario encompasses possible interventions with mitigation measures that are already available and are in practice. The mitigation measures and targets came from the Sustainable Energy for All (SE4ALL) initiative, SDGs of the United Nations, the Water and Energy Vision 2050 (WECS 2013), and low carbon economic development strategies (MOPE 2014). The major mitigation options identified are clean cooking technology, modern energy access, efficiency improvement in industrial processes, efficiency improvement in lighting, efficient mass transportation, renewable energy electricity generation, control on open biomass burning, intermittent aeration of rice field, animal waste management, waste management, and recovery of methane. The major emission factors for various activities were retrieved from Bond et al. (2004, 2013), EMEP/EEA (2013), IPCC (1996, 2006), Shrestha et al. (2013a), and Venkataraman et al. (2010). Details are tabulated in the Supplementary Materials.

Impact assessment method

Air pollution impacts human health, agriculture, and the environment at local, regional, and global scales. As such, it is necessary to quantify the effect of all emissions from all sources on the ambient pollutant concentrations, and associated impacts on human health, crop loss, and climate. In this work, “adjoint” coefficients, that quantify the sensitivity of a variable (e.g., an air pollution concentration impact metric) to emissions in 2 × 2.5° grids globally (see Henze et al. (2007)), from the GEOS-Chem atmospheric chemistry transport model (Bey et al. 2001) were combined with emissions to estimate the following: (1) population-weighted annual average fine particulate matter (PM2.5) and the maximum 6-month average daily maximum 1 h ozone (O3) concentrations, relevant for the impacts of these pollutants on human health; (2) 3-month average O3 concentrations across representative growing seasons for four staple crops (rice, wheat, maize, and soy); and (3) radiative forcing in four latitudinal bands (covering the Arctic, northern mid-latitudes, tropics, and southern hemisphere extra-tropics) due to emissions of each pollutant. Concentration–response relationships were then applied to estimate the air pollution-attributable premature deaths and air pollution-associated crop loss and to convert changes in radiative forcing to changes in temperature in each year following emission.

The combination of emissions and adjoint coefficients from the GEOS-Chem model were also used to assess the transboundary effects due to the transport of pollutant emissions from other countries. Outside Nepal, default emissions were used from the International Institute for Applied Systems Analysis (IIASA) ECLIPSE dataset (http://www.iiasa.ac.at/web/home/research/researchPrograms/air/ECLIPSEv5a.html, Stohl et al. (2015)) for all pollutants. In this study for Nepal, a grid of 2 × 2.5° resolution was used to analyze the emissions and their impact at a national scale. An overview of the models and methodology used for impact assessment in this paper is given in the supplementary materials and on the webpage of LEAP-IBC (Heaps 2017).

Economic evaluation

Economic evaluation of the impacts estimated for reference and policy scenarios in this study includes direct costs and benefits such as investment, cost of operation and maintenance, cost of resources, cost savings on fuel, carbon trade costs, and the economic value of lives and crop loss. It does not include indirect costs such as the cost of health services, income from tourism, employment generation, and ancillary productions. The costs are represented at 2005 constant price with social discount rate of 6%. The economic value of life was based on the value of a statistical life (VSL). Limitations of data mean that this has been derived by adjusting the ratio of Nepal’s GDP per capita to the EU average GDP per capita. It has been projected that Nepal’s best VSL estimate is about 83,000 USD with an uncertainty range of 41,000–124,000 USD (OECD 2012; World Bank 2015). The economic value of crops was estimated from a Food and Agriculture Organization estimate of producer price of each crop (FAOSTAT 2016).

Data sources for emission inventory

The steps followed in this study include the inventory development for emissions of particulates as well as gaseous emissions. A bottom-up approach was applied to develop the inventory at each activity level. In the REF scenario, the economic sectors were driven by GVA, which were retrieved from economic reports (MoF 2016; NPC 2014; NPC 2017). Another major driver of emissive activities is demography, which includes not only national population growth but also urbanization. These data were taken from CBS (2012, 2014). The current status of SLCP emissions in Nepal was derived within the LEAP-IBC modeling framework. The base year in the LEAP-IBC model is 2010. But, as we have passed the year 2016, the year 2015 has been taken as the base year to develop the emission inventory of SLCPs in Nepal. The major sources of energy and non-energy activities and technologies were retrieved from DOF (2016), FAOSTAT (2016), IRENA (2012), Manandhar and Dangol (2013), MOAD (2014), NEEP/GIZ (2012), Pradhan (2004), Shrestha et al. (2012), USEPA (2012), WECS (2010, 2014), and World Bank (2012). The sectors included in the study in addition to activity level data and emission parameters are given in the Supplementary Materials. The 2015 SLCP inventory is given in the “Emissions” section.

Uncertainty analysis

Uncertainties can arise from various factors such as inaccuracy in emission monitoring, lack of knowledge involving the emission factor, and activity data estimates. The uncertainty analyses for GHG emissions have also been recommended from the guidance by IPCC, and the method most commonly used in practice for uncertainty analysis is Monte Carlo simulation (IPCC 2000). In this study, the uncertainties are calculated at 95% confidence interval for energy and non-energy sectors using Monte Carlo simulation with 10,000 iterations. Sensitivity analysis has also been performed to examine the extent of variations in the results for BC, CH4, PM2.5, and GHG emissions due to input parameters.

Results and discussion

The results and figures presented in this chapter have been derived from LEAP (Heaps 2016).

Emissions

The emission inventory for 2015 was developed as the first step in results analysis. Table 1 shows the emission of different SLCPs and air pollutants from different sectors in 2015 from fuel and non-fuel combustion, as well as the selected chemical process.

Table 1 Major short-lived climate pollutants and particulate matter (tonnes) from different sectors in Nepal in 2015
Full size table

The results of both the REF and the POL scenarios, obtained from the LEAP-IBC modeling framework, are shown in Table 2. The table indicates that, with the policy intervention of different strategic measures (see Supplementary Materials), emissions of BC and PM2.5 can be greatly reduced in 2050 from their values in 2010. Similarly, emissions of CH4 and GHGs in 2050 can be reduced to near their values in 2010.

Table 2 Emissions in various years in reference and policy scenarios, in thousand tonnes
Full size table

From the perspective of sources, residential and commercial sectors are the prime sources of air pollutants. Thus, any strategy must place a strong emphasis on this sector. Meanwhile, in other sectors such as transport and industries, stringent pollution control regulations can help reduce pollution. The remaining sources either require low-emissive technological transformation or awareness to reduce emissive activities, such as reducing the open burning of wastes and biomass and waste reuse and recycle.

Impact analysis of scenarios for mitigation of SLCPs

The “Emissions” section described the overall emissions of various pollutants and climate forcers. However, the real measure that concerns everybody is their environmental impacts. This analysis does not include analysis of uncertainties on impact due to variations in BC, CH4, PM2.5, and GHG emissions. However, the uncertainty analysis for emission level for each pollutant is covered in the “Uncertainties in emissions and sensitivity analysis” section.

PM2.5 concentration

The national population-weighted annual average PM2.5 concentration was 47 μg/m3 in 2015, substantially above the WHO standard of 10 μg/m3 (WHO 2006). This value includes not only contributions from national emissions but also the influence of natural sources and transboundary emissions. The latter is estimated to make the larger contribution to national population-weighted PM2.5 concentrations, contributing almost 50% (Fig. 1). The reduction in PM2.5 concentrations has significant benefits, and these are discussed in the “Premature deaths avoided,” “Loss of crop yield mitigated,” “Impact on global temperature,” and “Economic evaluation of policy intervention” sections.

Fig. 1
figure 1

PM2.5 concentrations in Nepal in various years in reference and policy scenarios. REF, reference; POL, policy

Full size image

Premature deaths avoided

The most visible and significant impact of air pollution is on human health. Figure 2 indicates that reducing PM2.5 and O3 can help reduce large-scale health risks. Emissions from all sources in 2010 and 2015 resulted in an estimated annual air pollution health burden of 23,000 and 30,000 premature deaths, respectively; this accounts for around 1 per 1000 people. Given these figures, the REF scenario suggests that, if no mitigation strategies are implemented, the estimated air pollution-associated health burden increases to 50,000 premature deaths in 2030 and 109,000 premature deaths in 2050, with nearly 88% premature deaths due to PM2.5. In addition to increasing PM2.5 and O3 concentrations, these figures also reflect increases in, and aging of, Nepal’s population.

Fig. 2
figure 2

Premature deaths in Nepal in reference and policy scenarios. REF, reference; POL, policy

Full size image

With mitigation measures taken, the estimated total premature deaths from anthropogenic sources in Nepal in the POL scenario are reduced by 11,000 and 29,000 in 2030 and 2050, respectively. However, nearly 57% of premature deaths in the REF scenario are estimated to result from emissions outside Nepal. In the POL scenario, 77% of estimated premature deaths in 2050 are the result of transboundary air pollution because of a reduction in premature deaths due to national emissions. Thus, it is essential to take regional-level mitigation action to reduce transboundary emissions.

Loss of crop yield mitigated

The impact of ozone on four major crops (rice, wheat, maize, and soy) was assessed and was again mainly due to transboundary emissions. Thus, it is crucial that this issue of transboundary emission is addressed in the context of food security. The results indicate that crop losses can be greatly reduced in the POL scenario compared with the REF scenario (Fig. 3).

Fig. 3
figure 3

Crop yield loss in Nepal in reference and policy scenarios

Full size image

Impact on global temperature

One of the greatest concerns is the global warming due to pollutant release. The temperature increment is found to grow in future years, with SLCPs being the dominant factor. Thus, mitigation strategies for control of SLCPs are as important as those for GHGs to reduce Nepal’s contribution to global temperature increases. Figure 4 shows the temperature increment with reference to the temperature level in 2010. It can be seen that the temperature increment in the POL scenario before 2055 is higher than in the REF scenario. This is due to the dominant effect of cooling caused by aerosols. However, beyond this date, the reductions in CH4, CO2, and O3 precursors will ultimately reduce Nepal’s contribution to global temperature increases in the POL scenario. As major SLCPs decrease and GHG emissions are controlled, the temperature increment will be much slower than in the REF scenario, remaining below 2 mK in 2100.

Fig. 4
figure 4

Global average equilibrium temperature changes due to emissions in Nepal

Full size image

Economic evaluation of policy intervention

The cost of mitigation options is high in the energy sector, primarily because of the hydropower development. However, economic benefits due to a reduction in fuel imports, premature mortality, and crop losses are even larger. The costs of mitigation in the non-energy sector are higher than returns. But, owing to its huge contribution to emissions, and its impact, mitigation steps must be undertaken. At net present value, the overall investment and operating cost of implementing mitigation measures throughout the policy scenario sum to 21 billion USD, while the net benefit of 57 billion USD can be achieved. If the energy and non-energy mitigation strategies are implemented together, the benefits from the energy sector pay-off for the cost in the non-energy sector—and the overall benefit—are still positive. The net return is positive, with a benefit-to-cost ratio of 2.7. Thus, overall, the results indicate that the POL scenario is not only technically viable but also economically feasible.

The above economic analysis gives an important insight: mitigation options and activities not only are dependent on each other but also make the implementation process economically more viable if applied altogether. Thus, there must be inter-sectoral cooperation for implementation of the targets set to reduce SLCPs as well as GHG emissions. The economic analysis does not address the uncertainty ranges of SLCPs and the subsequent effects on other climate change benefits.

Uncertainties in emissions and sensitivity analysis

The variance propagation at 95% confidence interval of BC, CH4, PM2.5, and GHG emissions showed that the variance range reduces over time. The variance range in the POL scenarios narrows compared with the REF scenario fall for all four pollutants. In the POL scenario, the uncertainty range of BC reduces from 26–45 kt in 2015 to 5–9 kt in 2050. Similarly, that for CH4 narrows down from 0.8–1.3 Mt in 2015 to 0.7–1.1 Mt in 2050 and that for PM2.5 reduces from 200–280 kt in 2015 to 44–64 kt in 2050. The GHG uncertainty narrows down by little, from 29–43 Mt in 2015 to 38–49 Mt in 2050. These reductions in uncertainty range are due to switching to cleaner fuels with lower emission potential in the POL scenario. The sensitivity analysis for BC shows that the residential fuelwood emission factor and consumption make the major contribution to the variance in BC emission. The sensitivity analysis for CH4 shows that the emission is most sensitive to emission factors of livestock farming, residential fuelwood, waste, and rice cultivation. Similarly, the sensitivity analysis shows that the variance in the emission of PM2.5 is highly sensitive to the residential sector. Forest fires and emission factors of industrial coal and diesel consumption in transport are also major contributors to emission variance. Sensitivity analysis shows that in the REF scenario, the emission factor of livestock farming—including fermentation and manure management—makes the greatest contribution to variance in GHG emissions, followed by emission factors for CH4 and N2O of residential fuelwood.

Conclusions and policy implications

This study shows that the emissions and impacts of SLCPs in Nepal are significant with the major sources of BC and PM2.5 being biomass burning in the residential sector and fuel combustion in transport. The major source of CH4 emissions is agricultural activities followed by the residential sector and waste management. As anthropogenic activities increase, emission levels rise, contributing to adverse climate and air pollution impacts. If mitigation measures are taken in the POL scenario, 78% of BC, 78% of CH4, and 87% of PM2.5 emissions can be avoided in 2050 compared with the REF scenario. The national, population-weighted PM2.5 concentration of 47 μg/m3 in 2015 can be limited to 52 μg/m3 in 2050 compared with 80 μg/m3 in the REF scenario. Similarly, 29,000 premature deaths and 1.7 million tonnes of crop loss can be avoided annually by 2050 in the POL scenario compared with the REF scenario. The benefit-cost analysis indicates that there is a net economic saving of 36 billion USD (2005 constant price) if the strategic measures are undertaken in a timely manner. The impact on global climate due to emissions in Nepal could also be reduced by limiting temperature increment within 2 mK in 2100 in the POL scenario from near to 4 mK in the REF scenario. An estimated reduction of 58% can be achieved in 2100 if all the mitigation strategies are implemented.

Air pollution is not limited to a local area but has regional as well as global impacts. This paper suggests that emissions from the HKH region and beyond have a major influence on pollution levels and their impacts in Nepal, owing to the transboundary transport of pollutants. More than 50% of the PM2.5 concentration in Nepal was estimated to result from emissions outside the country. The situation in the rest of the HKH region is no different, as suggested by other studies as well (Kurokawa et al. 2013). This raises the need for regional cooperation among countries in the HKH region to act jointly in effective mitigation of SLCPs and reducing their impacts in the region. It is also essential that voices are raised in international organizations like UNEP and CCAC, requesting the necessary assistance in mitigating SLCPs, as the transboundary effects are at a much larger scale than the national-level effects.

These scenarios and results suggest that mitigation practices should be implemented as soon as possible, not only in Nepal but in other countries of the HKH region as well. The mitigation technologies are readily available, and supporting policies need to be devised and implemented. The inclusion and prioritization for mitigation of SLCPs in national policy are of utmost importance as their climate impacts are higher and short term in nature. Overall, coordinating this with similar and relevant strategies in the regional context can benefit the whole HKH region as well.

Several policy pathways could be followed for effective implementation of SLCP mitigation measures in Nepal that are so crucial for achieving several SDGs: from their role in reducing poverty to combating climate change, and engaging in adaption and mitigation. Developing an integrated approach to both air pollution abatement and climate change during the policy process is perhaps the most desirable pathway to maximize synergy, thereby making the public policy process more effective and efficient. Another pathway would be to build SLCP abatement policies on existing national development policies and initiatives. Integration of the economic costs of pollution into product pricing would incentivize consumers to make more informed choices, while at the same time creating pressure on producers to reduce their pollution footprint and adopt better practices. However, this calls for creating and supporting enabling conditions to confront a broad range of existing barriers to the design and implementation of national SLCP mitigation strategies. These include the need for a strong science–policy interface to raise awareness; data on pollution and its impacts; a dedicated regulatory institution with resources and capacity for effective implementation, monitoring, and enforcement; and changing the entrenched social norm and behavior of citizens.