An evaluation of solar powered irrigation as carbon offset projects ALEXANDER OLSSON

Licentiate Thesis in Chemical Engineering KTH – Royal Institute of Technology School of Chemical Science and Engineering Division of Energy Processes SE -100 44 Stockholm, Sweden

Copyright © Alexander Olsson, 2016 All rights reserved TRITA-CHE Report 2016:29 ISSN 1654-1081 ISBN 978-91-7729-057-5 Akademisk avhandling som med tillstånd av KTH i Stockholm framlägges till offentlig granskning för avläggande av teknisk licentiatexamen tisdagen den 30 augusti kl. 10:00 i seminarierumet plan 6, KTH, Teknikringen 42, Stockholm.

The thesis is intended for climate change policy makers and actors on the compliance and voluntary carbon markets. Abstract Carbon offsets have been developed as one tool to incentivise investments by developed nations in climate change mitigation activities in developing countries. The carbon offsets can be used towards the countries’ own mitigation targets but are also meant to benefit developing countries by providing a pathway to clean development. Photovoltaic water pumping (PVWP) technology is a solution to use PV for irrigation, which can be used to restore degraded grasslands and help farmers adapt to climate change. Restoration of degraded grasslands increases the production of grass and will therefore increase the amount of carbon in the soil, a process that may mitigate climate change. However, poor farmers often have limited access to irrigation technology and this thesis assesses how carbon offsets may bring revenues to increase adaption of PVWP technology in remote areas of the Chinese grasslands. PV modules can be used to mitigate climate change in different ways; the most common is to produce electricity to replace fossil fuel power capacity. The novelty of this thesis is that it assesses the alternative mitigation possibilities for the PVWP project proposed here. Further, consideration of water constraints that limit the applicability of the technology and a framework to assess the trade-offs between potential downstream water impacts and environmental co-benefits of the project add to the novelty of this thesis. Policy barriers for the project will also be considered. Used to restore severely degraded grasslands, PVWP projects show high carbon sequestration potential and successfully compete with grid electricity as carbon offset projects. A case is analysed and it shows that the carbon market could play a role in increasing the feasibility of PVWP projects. However, water issues make project

implementation very site-specific and some indicators to determine feasibility is proposed to be blue water availability, evaporation recycling ratio and water productivity. Water use must also be looked at with respect to climate, food and energy security, calling for a nexus approach to evaluate the project suitability. In May 2016, grassland management projects are excluded from the Clean Development Mechanism to the Kyoto Protocol, and this limits project implementation to the voluntary markets. Language: English

Keywords: carbon offset; carbon sequestration; clean development mechanism; climate change adaptation; CO2 emission reduction; desertification; solar power water pumping.

Avhandlingen är ämnad att läsas av beslutsfattare inom klimatområdet samt aktörer på de olika klimatkompensationsmarknaderna. Sammanfattning på svenska Klimatkompensation har utvecklats som ett verktyg för att stimulera industriländers investeringar i klimatprojekt i utvecklingsländer. Klimatkompensation kan användas för att nå industriländernas egna klimatmål men är också tänkta att gynna utvecklingsländer genom att tillhandahålla en ”ren” utvecklingsmöjlighet. Solcellsdrivna vattenpumpar (eng. photovoltaic water pumping: PVWP) är en teknik för att använda solceller för bevattning. Tekniken kan användas för att restaurera degraderade gräsmarker och för att hjälpa jordbrukare anpassa sig till klimatförändringarna. Restaurering av gräsmarker ökar produktionen av gräs vilket medför ökad mängden kol i marken, en process som kan mildra klimatförändringarna. Men fattiga bönder har ofta begränsad tillgång till bevattningsteknik och denna avhandling utvärderar hur klimatkompensation kan ge intäkter för att öka användningen av PVWP i avlägsna delar på den kinesiska slätten. Solceller kan användas för att mildra klimatförändringarna på olika sätt och vanligast är att producera el för att ersätta fossila bränslen. Det är därför viktigt att titta på alternativkostnaden för PVWP-projekten som föreslås här. Vidare begränsar vattentillgången projekten och ett ramverk för att tydliggöra avvägningar mellan vattenrelaterade problem och miljömässiga fördelarna med ett projekt är nödvändigt. Klimatpolitiska styrmedel sätter också upp vissa begränsningar för projekten. Om PVWP används för att återställa mycket degraderade gräsmarker, visar projekten hög klimatnytta och de kan framgångsrikt konkurrera med solel till nätet som klimatkompensationsprojekt. En fallstudie visar att klimatkompensationsmarknaden skulle kunna spela en viss roll för att öka antalet PVWP-projekt. Däremot gör vattenfrågan projektens geografiska plats viktig och indikatorer för att avgöra

genomförbarheten föreslås vara ”blåvattentillgång”, ”förångningsåtervinning” och ”vattenproduktivitet”. Vattenanvändningen måste också ses i förhållande till klimat, mat- och energisäkerhet, vilket kräver en nexusstrategi för att utvärdera projekten. I skrivande stund (maj 2016) är projekt rörande skötsel av gräsmarker exkluderade från mekanismen för ren utveckling (CDM) till Kyotoprotokollet och detta begränsar projekten till de frivilliga klimat-kompensationsmarknaderna.

List of appended papers:

I. Olsson A, Campana PE, Lind M, Yan J, 2014. Potential for carbon sequestration and mitigation of climate change by irrigation of grasslands. Appl Energy 136, 1145–1154. doi: 10.1016/j.apenergy.2014.08.025

II. Olsson A, Campana PE, Lind M, Yan J, 2015. PV Water Pumping for Carbon Sequestration in Dry Land Agriculture. Energy Convers Manag 102, 169–179. doi: 10.1016/j.enconman.2014.12.056

III. Olsson A, Grönkvist S, Lind M, Yan J, 2016. The elephant in the room – A comparative study of uncertainties in carbon offsets. Environ Sci Policy 56, 32–38. doi: 10.1016/j.envsci.2015.11.004

Contributions to the appended papers: Paper I I completed the work on developing the methodology described in the paper and verification of it with valuable input from Mårten Lind. Pietro Campana designed the system that we applied the methodology to and Jinyue Yan supervised the process and provided useful ideas and comments on the paper. Paper II Mårten Lind and I developed the idea of the paper and Pietro Campana helped with modelling in AquaCrop. Jinyue Yan supervised the process and provided valuable inputs on the paper. Paper III I completed most of the work for this paper with a lot of feedback from Stefan Grönkvist. Mårten Lind and Jinyue Yan also gave very useful inputs on the paper.

Related publications not included in this thesis Campana PE, Leduc S, Kim M, Olsson A, Zhang J, Liu J, Kraxner F, McCallum I, Li H, Yan J, 2016. Suitable and optimal locations for implementing photovoltaic water pumping systems for grassland irrigation in China. Appl Energy, Article in Press. doi: 10.1016/j.apenergy.2016.01.004 Olsson A, Lind M, Yan J, 2014. PV Water Pumping for Carbon Sequestration in Dry Land Agriculture. Energy Procedia 61, 1037–1041. doi: 10.1016/j.egypro.2014.11.1019 Yang J, Olsson A, Yan J, 2014. A Hybrid Life-Cycle Assessment of CO2 Emissions of a PV Water Pumping System in China. In the proceedings to The 6th International Conference on Applied Energy 2014, May 30 –June 2, Taipei, Taiwan. Campana PE, Olsson A, Li HL, Yan J, 2013. Economic analysis of photovoltaic water pumping irrigation systems. In the proceedings to The 5th International Conference on Applied Energy 2013, July 1-4, Pretoria, South Africa. Yan J, Gao ZY, Yu GJ, Campana PE, Liu JH, Zhang C, Yang J, Olsson A, Li HL, 2013. Demonstration and Scale-Up of Photovoltaic Solar Water Pumping for the Conservation of Grassland and Farmland in China. SIDA Project No.: AKT-2010-040.

Table of contents

1. Introduction ............................................................................................................... 1 1.1. Combating grassland degradation with solar power .................................................. 1 1.2. The United Nations Framework Convention on Climate Change ......................... 4 1.3. Climate change mitigation and other co-benefits of grassland restoration ................................................................................................................................. 5 1.4. Aim and research questions ........................................................................................... 6

2. Methodologies.......................................................................................................... 10 2.1. System level ..................................................................................................................... 10 2.2. What methodologies are used? .................................................................................... 11

3. Climate change mitigation and carbon offsets ........................................................ 13 3.1. Soil carbon sequestration .............................................................................................. 13 3.2. How to measure the increased soil organic carbon ................................................. 15 3.3. Additionality ................................................................................................................... 17 3.4. Leakage ............................................................................................................................ 18 3.5. Project and baseline scenarios ..................................................................................... 19 3.6. Comparing alternative carbon offset projects for the PVWP system .................. 21 3.7. Other limits to implementation of the carbon offset projects .............................. 26 3.8. Reflections on the feasibility of a PVWP carbon offset project............................ 28

4. Can water use for carbon sequestration be justified? ............................................... 29 4.1. Water use efficiency ....................................................................................................... 29 4.2. Water footprint............................................................................................................... 30 4.3. Evaporation recycling ratio .......................................................................................... 31 4.4. Water productivity ......................................................................................................... 32 4.5. The economic productivity of water .......................................................................... 33 4.6. Which indicators are relevant? ..................................................................................... 34 4.7. The nexus model – how do we value the use of water? ......................................... 35 4.8. Reflections on carbon sequestration, water and co-benefits .................................. 39

5. Policy aspects of PVWP project implementation ..................................................... 41 5.1. Leap frogging into a clean future ................................................................................ 41 5.2. Policy mechanisms for engaging developing countries in mitigating climate change ........................................................................................................................ 42 5.3. Uncertainties of carbon offsets and NAMAs ........................................................... 43 5.4. Non-permanence risk assessment ............................................................................... 46 5.5. The permanence period ................................................................................................ 47 5.6. Economic assessment ................................................................................................... 48 5.7. Water policy .................................................................................................................... 51

5.8. Reflections on policy issues and the economy of PVWP projects ........................ 52

6. Discussion ............................................................................................................... 54 6.1. The climate change mitigation potential .................................................................... 54 6.2. Water limitations ............................................................................................................ 55 6.3. The complexity of project implementation ............................................................... 56

7. Conclusions ............................................................................................................. 58

8. Future studies .......................................................................................................... 60

References .................................................................................................................... 61

Nomenclature CDM Clean Development Mechanism CERs Certified emission reductions CO2-eq CO2-equivalent DM Dry matter EF Emission factor ET Evapotranspiration FM Fresh matter GHG Greenhouse gas ha Hectare (10 000 m2) IRR Internal rate of return IPCC Intergovernmental Panel on Climate Change kWh kilowatt hour MRV Monitoring, reporting and verification NAMAs Nationally appropriate mitigation actions NPV Net present value PVWP Photovoltaic water pumping SCS Soil carbon sequestration SOC Soil organic carbon UNFCCC United Nations Framework Convention on Climate

Change VCS Verified Carbon Standard

Acknowledgements

Prof. Jinyue Yan, you have been my main supervisor and guide through this period. You have given me a lot of freedom to follow the tracks I wanted and always provided valuable feedback. Your experience from being Editor-in-Chief of Applied Energy has been invaluable when it comes to understanding how to stress the novelty and contributions of the research. Your experience from all over the world has given me so many influences and the possibility to meet interesting researchers. Thank you for that opportunity. I am very thankful to Mårten Lind PhD, my co-supervisor. You have provided me with the latest updates in the carbon offset field and always had very interesting opinions about my work. Stefan Grönkvist PhD, I have been teacher’s assistant in your course and towards the end we also wrote a paper together. I much appreciate how you have influenced my life and opened up new doors to systems thinking. Pietro Elia Campana, Chi Zhang, Hailong Li PhD and Jin Yang have been part of the same project team and you have been very good company on the conferences and field trips we have embarked on together. Prof. Per Alvfors has been the Head of Division of Energy Processes and I have much appreciated that you have given me the opportunity to partake in KTH Energy Platform. Mårten Larsson, Martin Bojler Görling, Johannes Persson, Yuting Tan, Mimmi Magnusson, Martina Wikström and Tomas Lönnqvist, thank you for making my time at Energy Processes both interesting and pleasant. I would also like to acknowledge the Swedish International Development Cooperation Agency (SIDA) and Swedish Agency for

Economic and Regional Growth (Tillväxtverket) for the financial support to the projects. Mattis Bergquist, you have been my mentor and listened to my problems, but also guided me through them. Courtney Adamson, my partner in life, you have helped me with illustrations and always been my best support.

1

1. Introduction

Solar photovoltaic water pumping (PVWP) systems for irrigation have been used to make increased production of grass, for example alfalfa, and some other crops possible in dry areas of western China (Yan et al., 2013). This thesis is part of a project called Demonstration and Scale-Up of Photovoltaic Solar Water Pumping for the Conservation of Grassland and Farmland in China. The project includes several pilot PVWP systems with the aim to integrate new PV technology, water-saving techniques and water resources management; to design robust user-friendly systems, and to develop a business model in order to increase agricultural output in poor areas of north-western China. This thesis contributes to the project through its analysis of the possibility to utilise carbon markets as a way to fund project implementation on a commercial scale. The analysis looks at the potential for climate change mitigation and the constraints put on the PVWP systems in water scarce areas. Ecological co-benefits to climate change mitigation are also analysed and the discussion is extended to different policies governing the carbon markets.

1.1. Combating grassland degradation with solar power

Things are changing on the Chinese grasslands. Previously nomadic people are becoming sedentary in a push from the government to prevent grassland degradation, believed to be caused by overgrazing (Fan et al., 2015). Following the modern transition from nomadic people to settlers, a shift in agricultural practices has been undertaken (ibid) and where sheep and goats used to roam the steppe, there are now irrigated fields of corn. Several small but empty towns are waiting for the resettlement of pastoralists along the road from Inner Mongolia’s capital city, Hohhot, to the village of Xilamu Rensumu in Wanlanchabu desert grassland area where the project Demonstration and Scale-Up of Photovoltaic Solar Water Pumping for the Conservation of Grassland and Farmland in China has a pilot plant.

2

The climate system that governs where humans can live, where we can grow food and find water for survival is changing and the change is unprecedented in modern times (IPCC, 2013: Summary for Policy Makers). The rate by which the climate is changing makes it dangerous to humans. Humans need time to adapt, to move to new places if the old ones are no longer habitable, and to develop new methods in order to survive in climates we know nothing about. This thesis deals with irrigation powered by solar photovoltaic (PV) technology, which is one way to manage human life in drylands and something that may find an increasing number of applications in a world with increasingly arid and variable weather. The PV powered system could also provide a way to allow a traditional nomadic lifestyle, while restoring and preventing further degradation of grassland areas. According to the United Nations Convention to Combat Desertification (UNCCD, 1994): “desertification means land degradation in arid, semi-arid and tropical sub-humid areas resulting from various factors, including climatic variations and human activities.” Climate models are predicting increased aridity in many places all over the world (Dai, 2011). At the same time, 2.1 billion people already live in drylands (UN, 2014) where the potential evaporation is much higher than precipitation, making water scarce and water management a life necessity. Grasslands cover between 30 % (Zhang et al., 2014) and 40 % (Hua and Squires, 2015) of China’s land area and the degradation of grasslands is estimated to induce a direct economic cost of CNY 54 billion, or almost USD 9 billion, per year (Akiyama and Kawamura, 2007). In April of 2015 Beijing was struck by what has been called the worst sandstorm in 13 years (Roberts, 2015). Increasing magnitude of sandstorms is only one of many consequences of land degradation in China, which is among the worst in the world (Hua and Squires, 2015). Desertification in China is debated and arguments for

3

manmade desertification (Akiyama and Kawamura, 2007) as well as climate change driven desertification (Wang et al., 2008a) are both convincing. Climate change induced desertification can however in many cases be exacerbated by human activities (Wang et al., 2008a). China has decided to go beyond ‘land-degradation neutral’ (Wang et al., 2013), and several land use policies to halt degradation, like the policy Grain for Green (Liu and Wu, 2010), have been implemented. Land restoration is also included as a contribution to the United Nations Framework Convention on Climate Change (UNFCCC) in China’s Intended Nationally Determined Contribution (INDC), where it is stated that they will continue to (China’s INDC, p. 10): “restore grassland from grazing land, to promote mechanism of maintaining the balance between grass stock and livestock, to prevent grassland degradation, to restore vegetation of grassland, to enhance grassland disaster prevention and farmland protection and to improve carbon storage of soil.” Several approaches to restoration of degraded grasslands exist, the simplest being excluding areas from grazing. This method has been quite successful in China and 60 million ha degraded grasslands have been included in the nationwide policy programme Returning Grazing Lands to Grasslands (Xiong et al., 2014). The method to exclude areas from grazing may, however, be more effective in areas with high precipitation (Xiong et al., 2014) and exclusion of grazing animals may have a high opportunity cost (Wang et al., 2011) and deprive pastoralists of their lifestyle (Yeh, 2005). In Tibet, fences to stop overgrazing could obstruct life of endangered wild migratory herbivores (Ji et al., 2016) and certain species of sedge grass in the Kobresia genus are driven out by other species in the absence of grazing animals (Qiu, 2016), and the result of that vegetative change could be a massive loss of carbon from the soil (ibid). Clearly, there is a need for other solutions to grassland degradation than fencing off land to exclude grazing animals, solutions that can benefit local herders and farmers while simultaneously mitigating the increasingly dry conditions on the grasslands.

4

Adaptation to climate change could be seen as the fundamental driver for the research conducted in this project and water is the main concern in the adaptation strategy of dryland people. Irrigation can adapt agriculture to short time dry spells in arid and semi-arid areas and increase yields and water use efficiency (Rockström, 2003). But long-term droughts can deplete groundwater resources and irrigation technology is rarely accessible to poor people (Kallis, 2008). Solar powered irrigation uses PV solar modules to generate electricity for pumps feeding water to an irrigation system. Grassland areas are often located far from the electric grid and off-grid systems are considered more cost effective than grid extension in most cases (Campana, 2013; Purohit and Michaelowa, 2008). The correlation of plant water demand and solar irradiance provides a unique feature and, at least on a monthly basis, generated electricity from the PV system and plant water demand therefore match well, making the intermittent nature of solar electricity less problematic for irrigation applications than for other applications (Campana et al., 2013a; Campana et al., 2013b). However, experience from India shows that the high capital cost prevents installation of PVWP systems despite many advantages to other pumping systems (Purohit and Michaelowa, 2008). These implementation problems are good reasons to look at additional co-benefits of the PVWP systems and the possible funding through carbon offset markets to overcome the barriers to implementation.

1.2. The United Nations Framework Convention on Climate Change

The United Nations Framework Convention on Climate Change (UNFCCC, 1992) is a legally non-binding treaty, i.e. an agreement under international law without legally binding targets on greenhouse gas (GHG) emissions for the Parties to the UNFCCC and without enforcement mechanisms (WMO, 2014). The Kyoto Protocol (UNFCCC, 1998) is a binding agreement within the UNFCCC and the countries listed in the Annex B to the Kyoto Protocol have committed to binding targets for emission reduction.

5

The Annex B countries are, with a few exceptions, the same as the Annex I countries to the convention (UNFCCC, 1992), and these two terms are consequently sometimes used interchangeably. The Annex I countries can be seen as the group that has the largest responsibility for the historical emissions and subsequently also the largest responsibility for reducing emissions. This is manifested in the emission reduction targets the Annex I countries have to meet. A system of how to comply with the targets is set up and this is called Kyoto accounting. The accounting principles include, among other things, the flexible mechanism called the Clean Development Mechanism (CDM), which allows Parties to buy Certified Emission Reductions (CERs) generated by climate change mitigation projects in countries without emission reduction targets. The CERs can then be used to offset GHG emissions. CDM is a completely voluntary mechanism, but is widely used since it is seen as a very cost-effective measure for the Annex I countries. This mechanism has the intent to help the countries that have agreed to specific emission reduction targets to achieve them, while assisting non-Annex I countries in achieving sustainable development (UNFCCC, 1998).

1.3. Climate change mitigation and other co-benefits of grassland restoration

Severely degraded lands have lost around 30 % of the soil organic carbon (SOC) stored in the ground (IPCC, 2006a). Soil carbon sequestration (SCS) can mitigate climate change by reducing the atmospheric content of CO2 and a major part of the carbon that has been lost through land degradation can be sequestered again, back to the ground (Lal, 2004a). This provides opportunities for carbon offset projects using SCS to mitigate climate change, while at the same time restoring degraded lands. In addition to the CDM, or the compliance market, there is also a voluntary carbon offset market. The voluntary offset market is primarily driven by proactive companies/organizations that want to measure and decrease their GHG emissions. The carbon offset credits that do not follow the CDM procedure can only be used on

6

the voluntary offset market and cannot be used by Annex I countries to meet their compliance under the Kyoto Protocol, but CERs produced in CDM projects can be sold on the voluntary market as well. PVWP projects could have a large potential as carbon offset projects on the voluntary market and potentially generate carbon credits that could make restoration profitable. However, land use, land-use change and forestry (LULUCF) activities under CDM have been limited to afforestation and reforestation, excluding other LULUCF activities like grazing land management. The discussion on how to include additional LULUCF activities was initiated in 2012 at the Conference of the Parties (COP) 17, the climate meeting in Durban (UNFCCC, 2012a). Some Parties to the UNFCCC have cautioned that the potentially large impacts of CDM LULUCF activities would be used as a loophole for Annex I countries to postpone their emission reduction commitments, and instead invest in these potentially much cheaper mitigation projects (Boyd et al., 2008). Nonetheless, the discussion on how to include additional LULUCF activities continues and if new methods are accepted it will be a huge opportunity for restoration of degraded grasslands. Important obstacles to overcome will then be to find innovative methodologies to adequately monitor, verify and report carbon sequestration in grassland restoration, without high transaction costs, in the near future.

1.4. Aim and research questions

The technology of the PVWP system and dynamics of water demand and solar irradiation, have all been covered by Campana (2013) as part of the same research project. Modelling of groundwater table variations and hydrology has been conducted by Gao et al. (2013), and their model was further extended by Zhang et al. (2014a). The overall aim of this thesis is to assess the potential to mitigate land degradation and climate change by PVWP for irrigation of grassland

7

and farmland in China, and more specifically to assess the potential to use the technology in carbon offset projects. A broad range of analysis tools to assess not only the mitigation potential but also adaptation to climate change and sustainable development benefits of the PVWP system have been used. The assessment also takes water constraint of dryland agriculture into consideration. New perspectives on water use, climate change mitigation and adaptation, energy and food security, as well as moisture recycling in a so-called nexus approach are introduced. Furthermore, efficiency improvements and rebound effects are discussed in relation to carbon offset projects. This is included not as a critique against technological solutions for climate change and water scarcity mitigation, but to show that policy issues are equally important and to caution against overly optimistic attitudes towards technical solutions. Several scholars have estimated the carbon sequestration potential in the world’s grazing lands (Henderson et al., 2015; Conant et al., 2001) and some have more specifically published work on SCS in grasslands in China (e.g. Wang et al., 2011; Lal, 2002). On the other hand, a few researchers in the engineering field have estimated the climate change mitigation potential when using PVWP systems instead of diesel fuelled generators for irrigation (e.g. Ould-Amrouche et al., 2010). However, there is a need for a methodology combining the two approaches, both SCS and emission reduction, to utilise the full potential of the PV system to mitigate climate change. There is also a need to investigate the opportunity cost of the PV system when used for carbon offset projects. At the time of writing, only three irrigation projects can be found in the CDM project database (UNFCCC, 2014a) and although some scholars have investigated the possibility to use PVWP projects as CDM projects (e.g. Purohit and Michaelowa, 2008) no one has considered the possibility to use PVWP to restore degraded grasslands and gain credits for carbon sequestration instead of fossil fuel substitution. Furthermore, the consequences of using water for carbon sequestration have been analysed by Rockström et al. (2012) among others. However, the complexity of the impact from carbon sequestration projects on water flows, food and energy security is not recognised in the literature and a model for valuing

8

different aspects and trade-offs in these projects is needed. Linked to the project implementation is the policy around LULUCF activities. The debate on whether to include additional LULUCF activities other than afforestation and reforestation in the CDM or not is on-going (UNFCCC, 2014b) and researchers have a responsibility to suggest scientifically sound policy advice on the topic. Figure 1 states some research questions that have been important when writing this thesis. The general outcomes from the research questions and where in the thesis they are dealt with has also been included in the figure.

Figure 1: Thesis structure based on research questions and outcomes. All the outcomes have the aim to help the analysis of the feasibility of PVWP projects as carbon offset projects.

9

PVWP systems for irrigation of grassland and farmland in China, show great promise to improve pastoralists’ incomes and livelihoods whilst at the same time restoring degraded grasslands (Campana, 2013). Since CERs have the potential to provide extra incentive to invest in a PVWP system it could be a feasible way to make the system economically viable. It has therefore been the aim to answer some of the pressing issues on how to value concerns and benefits of these kinds of carbon offset projects against each other. Hopefully this thesis will provide some insight to local assessments and policy makers on the suitability of PVWP in restoration of degraded lands.

1.5. Thesis outline

The thesis starts with a short discussion of the methodologies used (section 2). Section 3 then goes on to deal with the potential to mitigate climate change and how a PVWP project could be used as a carbon offset project. The implications of water use for carbon sequestration are discussed and additional environmental benefits of the restoration process are included in a water-food-energy-climate nexus model that is introduced in section 4. In section 5 barriers and opportunities regarding the use of PVWP projects as carbon offset project are introduced. The bias towards other activities than LULUCF activities in the CDM and the uncertainties of different mitigation options within CDM are also discussed. Section 6 summarises the discussion and section 7 concludes the thesis. Finally, section 8 suggests scope for further research. Some text has been underlined in the thesis to highlight interesting consequences and conclusions.

10

2. Methodologies

“You can’t study climate or hydrology or fertility without studying all of these issues, they don't exist in isolation.”

- John D. Liu, Chinese American filmmaker and ecologist (Savory Institute, 2014).

2.1. System level

In the assessment of the potential to mitigate grassland degradation and climate change, there are several aspects that can be deemed important. Knowledge from soil science, water science, meteorology, as well as climate change research both from a social, economic, political and natural science point of view has to be incorporated to understand the system. I have taken a multidisciplinary approach and tried not to delimit the research questions to quantifiable problems, something otherwise quite common in the engineering field. Studying at Division of Energy Processes at KTH has opened up the world of multidisciplinary research for me personally and made me critical to ‘tech-fix’ solutions, without reducing my respect for technology and engineer’s role in society. The fact that this thesis has been written within a project set on a certain technology to solve a socio-ecological problem, could be seen as problematic. Ziman (2002) writes that this way of conducting science is “production of knowledge with clearly foreseen or potential uses” something he calls “instrumental research”, and goes on: However, science also has valuable non-instrumental social functions. Open-ended, disinterested research enriches society with influential, trustworthy, general knowledge. The thesis is not an open-ended assessment of how to create ecologically, socially and economically sound carbon offset projects in the degraded Chinese grasslands. It is instead an investigation of how a specific technology would perform as a carbon offset project,

11

following a bottom-up approach. Questions regarding better, cheaper or more socially accepted ways (if they exist) to adapt grassland communities to climate change and mitigate global warming will therefore not be answered. Despite the limitations of project-induced research, I have tried my very best to be “imaginative enough to explore all aspects of the natural world” something Ziman suggests that a researcher should do (ibid). The choice of system level may determine the results of an assessment. The most striking example in this thesis could be whether the importance of vegetation for rain production (van der Ent et al., 2010) is included in an analysis or not (see section 4 Can water use for carbon sequestration be justified? of this thesis). Another interesting system aspect is how domestic food security should be valued compared to trade with virtual water. Import can make it possible to consume more water-demanding products than would have otherwise been possible in that region. It may also allocate production to a region where the good can be more efficiently produced in terms of water use per product. At the same time, some products are important to the domestic food industry. In the Chinese case it may for example be deemed more important to produce alfalfa for animal fodder domestically, than to rely completely on import from the U.S.A, which is the situation today (Bean and Wilhem, 2010), despite ‘comparative (water) advantages’ in the U.S.A. It is part of every countries national food security considerations to be more or less reliant on foreign trade.

2.2. What methodologies are used?

Instead of monetizing environmental benefits and ecosystem services*, I start with the assumption that human life in the Chinese grasslands is worth protecting and I do not try to put a quantitative or monetary value on the human civilization in this area. PVWP can contribute to making this life possible and attractive. Following this

* Ecosystem services are the benefits people obtain from ecosystems, e.g. pollination by insects.

12

assumption I have taken the approach to describe the environmental benefits and concerns to gain knowledge and awareness of them. To estimate the climate change mitigation benefit from restoring degraded grasslands using PVWP for irrigation, a straightforward engineering approach is used. The approach is simple but gives a good estimate of the potential of this restoration method and this methodology is mainly developed in Paper I. Paper II builds on experience from field tests with PVWP pilots in Qinghai and Inner Mongolia provinces in China and goes into several other issues needed to justify water use in dryland agriculture. A general model is constructed from the observations made in the field. Water is almost inevitably scarce in the degraded drylands and water use for one activity should be compared with alternative uses of water. To assess if water use for restoration can be justified, PVWP is analysed from a nexus perspective taking into account several environmental benefits and concerns that can come about from the restoration process. Paper III finally analyses the policy landscape of carbon offsets and argues for including additional LULUCF activities in the CDM. Paper III discusses the potential to mitigate climate change in the LULUCF sector and compares the uncertainties between different project categories within the CDM. Both Papers II and III use literature reviews and general systems thinking to arrive at important results for implementation of PVWP projects. The complex nature of the problems dealt with in both Papers II and III makes it hard to apply one single methodology and come to a clear-cut solution at the other end. Instead, arguments are presented, discussed and weighed against each other before leading to a conclusion of how the problems may be handled.

13

3. Climate change mitigation and carbon offsets

“Small changes in the soil organic carbon stock could result in significant impacts on the atmospheric carbon concentration.”

- Stockmann et al. (2013)

3.1. Soil carbon sequestration

The soil carbon pool constitutes the largest biologically active carbon pool on earth (Stockmann et al., 2013) and due to land-use change from, for example, forest or grassland to cropland, carbon is emitted to the atmosphere from the soil in the form of CO2, contributing to the increase of GHGs in the atmosphere. Somewhere between 42 and 78 Gt (109 tonne) of carbon has been lost from soils globally due to active agricultural management and unintentional land degradation (Lal, 2004a). Part of the carbon has probably been buried in sediments but most of it has been oxidized to CO2 and contributes to global warming (Stockmann et al., 2013). The SOC pool is dynamic and plant and root litter is decomposed at different rates depending on the complexity of the litter. Roots also provide sugars, amino and organic acids, so called root exudates, which are rapidly decomposed by microorganisms in the soil. SOC is commonly divided into three pools: a fast pool that decomposes in days to years, a slow pool that decomposes during years to decades, and a passive pool where decomposition takes decades to a millennium (Stockmann et al., 2013). Several management methods can increase the levels of SOC in soils. Restoration of degraded lands, no-till agriculture, enhancing crop rotation complexity, fallow, or growing perennial crops instead of annuals are methods that can increase SOC stocks (ibid). But the increase rates vary across the globe and system aspects such as increased use of fossil fuel, enteric methane emissions etc., should be included to calculate the net climate change mitigation benefit of such measures (ibid). One strategy to increase SOC stocks is to plant nitrogen fixing species such as alfalfa, which leads to increased levels of microbial nitrogen in

14

the soil and increased SOC stocks (ibid). However, this practise may increase emissions of N2O. It is therefore important to assess the net contribution to global warming using this strategy (Henderson et al., 2015). A large part of the carbon lost from the ground can be sequestered back into the soil and the global technical potential is 4.9-5.3 Gt CO2 per year for several decades according to the 4th assessment report by the Intergovernmental Panel on Climate Change (IPCC) (IPCC, 2007: chapter 8). However, IPCC (ibid) estimates the economic potentials from cropland management to 750 Mt (106 tonne) CO2-equivalents (CO2-eq) per year, from restoration of cultivated organic soils to 250 Mt CO2-eq per year and from grazing land management to 180 Mt CO2-eq per year at a price of USD 20 per tonne CO2-eq. The total mitigation potential from the agricultural sector under this price is 1.5-1.6 Gt CO2-eq per year by 2030, and 89 % of this potential is due to soil carbon sequestration (SCS). These activities could in many cases increase food security and water quality, and thus be a no-regret strategy for climate change mitigation (Lal, 2004b). Since increased levels of SOC contribute to more soil water and increased resilience against drought (Gordon and Enfors, 2008: Figure 3.4a), SCS can also act as an adaptation strategy to climate change. What this means is that there is an underutilised win-win strategy in SCS that can mitigate climate change with around 4 % of the global fossil fuel emissions per year (compared with 34.5 Gt CO2-eq global fossil fuel emissions in 2012; Olivier et al., 2013). The Chinese grasslands store 9-16 % of all the grassland carbon globally. If emitted to the atmosphere this carbon would be equivalent to 11 and 150 Gt CO2 from the vegetation and the soil carbon pools respectively (Ni, 2002). There is, however, significant potential to store more carbon in the Chinese grassland soils (Lal, 2002). Over 5 Gt CO2 has been lost from Chinese grasslands due to unsustainable land management (Wang et al., 2011), which could be reversed, and the total technical potential for SCS in China has been estimated to 40 Gt CO2 at a rate of 0.82 Gt CO2 per year (Lal, 2002).

15

3.2. How to measure the increased soil organic carbon

The standard method for measuring SOC is a chemical analysis using dry combustion, i.e. weight loss after combustion of carbon in the soil (Stockmann et al., 2013). This requires taking a soil core sample from several locations since SOC density varies a lot with landscape position. While methods for sampling and future re-sampling of SOC accumulation rates are easy and well established, they can be very costly for a small sequestration project seeking to verify its mitigation benefit under a climate change policy scheme (Conant, 2012). A combination of ex-situ methods like the one described and in-situ methods, such as near- and mid-infrared spectroscopy, could be a cost effective alternative for monitoring of SOC stock increases (Stockmann et al., 2013). A simpler approach than the methods described above is to model SOC changes. This requires gathering of data on biomass input, climate and management regime and the availability of long-term data sets for calibration of the model (Stockmann et al., 2013). Common models are CENTURY and RothC. RothC requires input data of rainfall, air temperature, open pan evaporation, clay content, the input of carbon from organic matter to the soil, an estimate of the decomposability of the specific organic matter and information about the vegetation cover (Farina et al., 2013), all of which could be quite easily monitored under a land management project. Model predictions of SOC levels in soils have been shown to be within the uncertainty range of measured levels (Smith et al., 2012). Modelling of SCS requires biomass input data from real operations. In Paper I it was instead desired to produce an estimate before the project is taking place and it was necessary to find a method omitting some of the input data required for modelling SOC dynamics. The feasibility analysis proposed in Paper I relies on default values derived by the IPCC. The input is based on the soil type, the land use relative to the native conditions (FLU), management practice under the carbon sequestration project (FMG) and organic matter input (FI). Default values for these factors are specified for each climate zone and they

16

are multiplied with the SOC reference value (SOCREF) to obtain the difference in equilibrium value between the two states. The SOC for a certain management regime is then obtained with equation 1. The transition to the new equilibrium is assumed to take 20 years. Verification of the default method is necessary since it is very uncertain and the study by Wang et al. (2011) proved very useful for Chinese grasslands in this respect. SOC = SOCREF · FLU · FMG · FI (1) where: SOC = the carbon stock [tonne carbon per ha for 0–30 cm depth] SOCREF = the reference carbon stock [tonne carbon per ha for 0–30 cm depth] FLU = stock change factor for land use or land-use change type [dimensionless] FMG = stock change factor for management regime [dimensionless] FI = stock change factor for input of organic matter (applied only to improved grassland) [dimensionless] Afforestation and reforestation projects do not suffer the same problems as SCS projects since simplified methods to monitor the tree growth, like measuring the diameter of the tree at breast height, have been developed. However, soil carbon is not visible to the naked eye and the use of models could render some scepticism. Regional values for the default input parameters could be a way to increase accuracy on project level without imposing cumbersome measures on the project developers. But to develop regional default values will take time and delay project implementation. A simple methodology developed by the BioCarbon Fund and approved by Verified Carbon Standard (VCS) for offset projects can be used for project on the voluntary market (VCS, 2011). RothC is recommended for modelling of SOC stock increase and instead of calibration of the model, VCS allows studies that show the appropriateness of using the model in the ‘agro-ecological zone’ (ibid). Default values for decomposition parameters can also be used

17

in the absence of real data. A simplified version of the RothC model has been developed and validated under VCS for use in sustainable agricultural land management projects (World Bank, 2013).

3.3. Additionality

When a project activity decreases emissions of GHGs or sequesters carbon in a sink, this is measured relative to a baseline scenario, also called a business as usual scenario. All carbon offset projects have to be additional, and additionality may be proven if it is clearly shown that the project scenario differs from the baseline scenario. Baseline scenario analysis faces an inherent problem: it is future telling and there is no way to verify if the project baseline is in fact the correct one. Ould-Amrouche et al. (2010) replaced diesel generator pumping systems with PVWP systems. In this case, diesel generators constitute business as usual, or the project baseline scenario. Thus, it would have to be proven that this is common practice and that PVWP systems are not. The CDM rulebook (2014) lists the following barriers that may prove additionality:

1) Investment barrier 2) Access-to-finance barrier 3) Technological barrier 4) Barrier due to prevailing practice 5) Other barriers

Only one barrier has to be present to prove additionality. Additionality for land restoration should mean that similar activities are not common practice and that the incentive that carbon credits can give will have a direct and deciding influence on the feasibility of the project. Ultimately a methodology developed for a carbon sequestration project must be able to convince the buyer and the controlling agency (under the CDM the controlling agencies are called Designated Operational Entities) that mitigation of climate change is actually taking place. A tool to help LULUCF projects in this work is

18

Global Land Degradation Information System (GLADIS; Nachtergaele et al., 2011) and other land monitoring tools that can assist in assessing trends of land-use change in the area. Furthermore, if the investment in a PVWP system leads to revenues from grass production it could be questioned if the project is additional. But land restoration is rarely profitable and carbon offsets could assist in restoring lands that would give low revenues from grass production and sales.

3.4. Leakage

Leakage under CDM is emissions of GHGs that can be directly attributed to the project but occur outside the system boundary of the project (UNFCCC, 2006b). Leakage for the PVWP system may include fossil fuel combustion during construction of the water well and emissions of N2O due to increased biomass production and increased soil nitrogen. Therefore, leakage is included in the mitigation benefit calculations throughout this thesis. According to a lifecycle analysis, emissions arising from production of PV panels and other components are lower than 50 g CO2-eq per kWh (Peng et al., 2013). This is small compared with emissions from grid electricity in China, which are around 900 g CO2-eq per kWh depending on province. Tracking lifecycle emissions back through the production and supply chains is cumbersome and associated with huge uncertainties. Given the low lifecycle emissions this part is neglected in the analysis, which is in line with the CDM methodology AMS-I.D.: Grid connected renewable electricity generation, version 17.0 (AMS is short for approved methodology for small scale projects) (UNFCCC, 2012b). Several types of CDM methodologies have neglected emissions from construction of infrastructure. In biogas projects, e.g., emissions from construction of the digester are omitted. Zurbrügg et al. (2012) compared emission reductions from organic waste treatment projects compared to landfilling. They used the prescribed CDM methodologies and compared it with a lifecycle analysis and showed that the CDM methodologies underestimate the

19

projects climate change mitigation benefit. Construction of biogas reactors contributed with emissions equivalent to 0.3 % of the baseline emissions from the landfill (Zurbrügg et al., 2012). It may therefore be reasonable to simplify a methodology on PVWP and omit the lifecycle analysis and this was done in the methodology in Paper I. It is important to point out that indirect emissions from production of components in a carbon offset projects are neglected both in baseline and in project emission estimate and it could therefore be argued that they, at least to some extent, cancel each other out. Wu et al. (2014) showed that low levels of nitrogen are limiting SCS in their study area in Inner Mongolia when restoring degraded grasslands. This would motivate the use of legumes to increase productivity and SCS. Increased microbial nitrogen may however increase N2O emissions, which is then attributable to the project activity and should be considered leakage. Paper I shows that this leakage is potentially large, and Henderson et al. (2015) concluded that N2O emissions may limit climate change mitigation using legumes in parts of the world’s grasslands. Furthermore, there could be an increase in the amount of grazing animals supported in the project area due to increased biomass production. But the increased production of forage could very well lead to a decrease in the grazed area and more land being set aside for conservation instead. This could vary from project to project and an assessment of the situation should be a part of a project feasibility analysis.

3.5. Project and baseline scenarios

The methodology developed in Paper I uses a novel baseline scenario. The decrease in GHG emissions relative to the baseline scenario is the climate change mitigation benefit of the project activity and depending on what the baseline is the climate change mitigation benefit will be different. The baseline scenario used by Ould-Amrouche et al. (2010), where a PVWP system replaces diesel-powered generators, would calculate the diesel used by the pumps

20

under baseline and from that achieve the reduction in GHG emissions. This is a straightforward replacement of fossil fuels and would be easy to monitor under existing CDM methodology. However, no such CDM project exists and there is reason to doubt the feasibility of this type of projects, something that will be discussed in section 3.6 Comparing alternative carbon offset projects for the PVWP system of this thesis. When constructing baseline scenarios for grid connected energy projects a detailed analysis of the operating margin must be made, i.e. an analysis of the existing power producing units and their GHG emissions. All the future plans of power generation projects in the area must also be taken into consideration, called the build margin, and from this information it is possible to estimate the GHG emissions arising from grid electricity for the baseline in absence of the CDM project. If CDM projects are assumed to substitute grid electricity, it is then possible to calculate the amount of CERs that the project should be rewarded with using the combined margin, a combination of the operating and build margin. Baseline is in this way determining the amount of CERs that will be produced by the projects. Purohit and Michaelowa (2008) studied PVWP systems and considered both substituting diesel generators and grid electricity, thus a grid emission factor (EF) had to be determined. The EF is an estimate of the GHG emissions connected to grid electricity production, given the different varieties of fuels used, and if the PV system replaces grid electricity the EF determines the baseline scenario. When proving additionality in the grid-connected case, baseline should constitute grid-connected pumps. Baseline in both examples above already includes the water pumping system itself, and the project scenario only alters the technology with which electricity is generated. However, irrigation to restore degraded grassland is usually not common practice and there is no reason why using a diesel generator should be assumed as the only possible baseline scenario in the carbon offset project described here. Severely

21

degraded grasslands have undergone major long-term loss of productivity and vegetation cover, due to severe mechanical damage to the vegetation and/or severe soil erosion and this has depleted the SOC pool (IPCC, 2006a). This fact is used to form the baseline approach proposed in Paper I, instead using the increment of the SOC pool due to the project activity. In addition to this mitigation benefit, the excess electricity produced by the PV system can be used to substitute grid electricity to increase the total amount of carbon credits being produced.

3.6. Comparing alternative carbon offset projects for the PVWP system

Three alternative carbon offset projects are possible to distinguish from the discussion in section 3.5: A1: PVWP project that will replace a diesel powered pumping system. A2: PVWP project that will restore degraded land using irrigation and reseeding with legumes. A3: a project where the PV system is instead used to produce grid electricity. A3 is used as a reference scenario to see if PVWP makes sense as a carbon offset project, and we would like to know which one of these three projects give the highest mitigation benefit. The useful power output from the PVWP system, i.e. the output that can produce mitigation benefit, is limited by the water demand in A1 and A2, and simply connecting the PV system to the grid, like in A3, will always supply more electricity to the grid than the stand-alone PVWP system will supply to the pump. However, it may be possible to achieve additional mitigation benefits under the configurations in A1 and A2, and it is therefore interesting to compare these two projects to A3.

22

A1 compared to A3: Replacing a diesel powered pumping system A1 considers a project where the existing way of irrigation, i.e. the baseline, is to use a diesel powered generator to drive a pumping system. The GHG emissions from the baseline scenario are given by the EF for the diesel pump (DP), EFDP, and the project sets out to replace the diesel system with a solar powered PVWP system. To make A1 a viable CDM project, the total baseline emissions from the diesel pump must be larger than or equal to the total electricity production from the PV system, which can be expressed with the inequality in equation 2. EFDP · kWhDP ≥ EFgrid · kWhgrid (2) where: EFDP = emission factor for the diesel pump [g CO2 per kWh] kWhDP = electricity needed by the diesel pump [kWh] EFgrid = emission factor for the electrical grid [g CO2 per kWh] kWhgrid = total amount of electricity delivered to the grid by the PV system [kWh]

Considering that kWhDP < kWhgrid since the PV system will produce electricity even when there is no demand for water, EFDP > EFgrid is the only possibility for equation 2 to hold. EFgrid varies for different regions. In some provinces in North China EFgrid = 931 g CO2 per kWh and as an average in India EFgrid = 900 g CO2 per kWh (Purohit and Michaelowa, 2008). For small diesel-driven generators (<15 kW) powering a water pump, the load factor is around 50 % and EFDP = 1400 g CO2 per kWh pump work (UNFCCC, 2006a). The PV system will generate more electricity than needed for most of the time. In fact, the PV system analysed in Paper I produces more than twice the electricity needed for pumping water (Olsson et al., 2014) and EFDP must therefore be around twice the value of EFgrid, i.e. EFgrid = 700 g CO2 per kWh, for A1 to be feasible compared to A3. No province in China fulfils that requirement (Chinese DNA, 2010). Thus, to maximise CERs production it would be more beneficial to supply renewable electricity to the grid with the PV system than to substitute

23

diesel generators and A1 is not a feasible carbon offset project in China under the current EFs. To maximise the mitigation benefit of the carbon offset project A1, the excess electricity generated by the PV system may be supplied to the grid to gain extra CERs. We can call this alternative A1*, which can be described with equation 3. EFDP · kWhDP + EFgrid · kWhexcess ≥ EFgrid · kWhgrid (3) kWhexcess = the excess electricity produced by the PV system and delivered to the grid [kWh] This would utilise the fact that EFDP > EFgrid while at the same time using the excess electricity to generate CERs, improving the mitigation benefit compared to A3. However, despite this possibility and a large potential for PVWP systems as CDM projects (Purohit and Michaelowa, 2008) not a single project has been implemented. Only three irrigation projects can be found in the CDM project database (UNFCCC, 2014a). Two of the projects both suggest substituting diesel with biomass fired combined heat and power production, making the electricity available on demand compared to the variability in power output that characterises a PV system. The third project introduces a micro irrigation system to reduce the electricity need for pumping and thereby claims a reduction in CO2 emissions. Hence, the small increase in mitigation benefit achieved in A1* (see Figure 2) compared to A3 is probably not worth the trouble of both substitution grid electricity and replacing off-grid diesel powered pumps, and thus PVWP systems as CDM projects under the current CDM regime do not seem to be attractive. A2 compared to A3: Soil carbon sequestration through land restoration using irrigation and reseeding with legumes A2 is the system described in Paper I and it is a land restoration project using irrigation to improve the production of nitrogen fixing legumes. Also with this baseline is it possible to assess the opportunity cost of the PV system in terms of climate change

24

mitigation compared to A3 and the inequality that has to be fulfilled to make the project feasible can be described by equation 4. SCS ≥ EFgrid · kWhgrid (4) where: SCS = the soil carbon sequestration from irrigation (including leakage) In this case we look for the answer to the question ‘from a climate change mitigation point of view, would it be better to use the PV system to replace grid electricity or to restore degraded grasslands?’ The answer will vary with the EF of the local electric grid. In the area we analysed in Paper I, Qinghai province in China, the EF is calculated with the combined margin approach used in CDM to 931 g CO2-eq per kWh. Since the system analysed in Paper I produces 4678 kWh per year and estimated SCS is 4.28 tonne CO2 per year, the inequality does not hold: 4678· 931· 10-6 = 4.36 tonne CO2 per year > 4.28 tonne CO2 per year This means that it would be better to use the PV system to replace grid electricity than to use it for a stand-alone irrigation system, if the sole objective is to maximise the climate change mitigation benefit. Instead we may analyse the alternative A2* described by equation 5, where the excess electricity is supplied to the grid, improving the mitigation benefit of A2. This is achieved by a mobile PV system that can be transported and connected to the grid after being used for irrigation during May-September. SCS + EFgrid· kWhexcess ≥ EFgrid· kWhgrid (5) In A2*, the mitigation benefit is increased substantially and it could therefore be a realistic alternative for a carbon offset project. The results are shown in Figure 2.

25

Figure 2: Comparison of the annual mitigation benefits from five different carbon offset projects using the same PV system. The five projects are described in the text. While the case-specific calculations carried out here and in Paper I will change considerably with location, the case shows that utilising the excess electricity is essential for the PVWP carbon offset project to be feasible compared to A3. If no grid connection is available at the project location, it would require the PV system to be transported and a mobilized system with the PV array mounted on a trailer and moved during the months without irrigation can be a solution. The possibility to set up a local mini grid could also be possible but may require additional investments and be part of a larger carbon offset project, including electrification of rural areas. This would make it possible to use excess electricity on-site and transportation of the PV system would not be necessary.

1.84

4.28 4.36 4.99

6.63

0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

A1 Dieselsubstitution

A2 SCS withirrigation and

legumes

A3 Gridelectricity

substitution

A1* Dieselsubstitutionwith excesselectricity to

the grid

A2* SCSwith excesselectricity to

the grid

To

nn

e C

O2 p

er

ha

Annual mitigation benefit

26

3.7. Other limits to implementation of the carbon offset projects

In the LULUCF sector, the technical mitigation potential refers to a biophysical possibility to mitigate climate change (IPCC, 2014: chapter 11). This potential is interesting when looking at different mitigation technologies from a theoretical point of view and provides innovators with the physical boundaries for mitigation potentials. The economic mitigation potential gives a more realistic picture of the mitigation potential that is actually achievable. A price on carbon is typically set when determining the economic mitigation potential and the economic feasibility to mitigate climate change is estimated for this price. Economic potentials are subsequently expected to be lower than the technical potential (IPCC, 2014: chapter 11). The market mitigation potential is taking into consideration also, inter alia, cultural and institutional barriers, and is normally expected to be lower than the economic potential. Figure 3 illustrates the different barriers and potentials for climate change mitigation. Since several forest and agriculture mitigation options sometimes overlap, it is difficult to estimate economic and market mitigation potentials in the LULUCF sector. This interdependency of different land use options also creates a complex web of opportunity costs. In dry areas where biomass is limited, there is competition over the use of crop residuals. If left in the field crop residues contribute to SOC, but could alternatively provide animal feed or fuel for cooking and heating. Introduction of renewable electricity in the agricultural sector increases the complexity of estimating these opportunity costs.

27

Figure 3: Different barriers to achieving the maximum mitigation potential that is physically possible exists in society and can be economical or social and institutional. Figure adapted from IPCC (2014: chapter 11). The economic feasibility of the concept in Paper I relies on sales revenue from alfalfa production (Campana et al., 2013a). Skinner (2008) showed that biomass removal from mature pastures limits SCS and some of the alfalfa would have to be left to decompose in the field. However, degraded grasslands do not constitute mature pastures. Furthermore, root litter also contributes to SOC accumulation (Stockmann et al., 2013) and Su (2007) harvested hay three to four times per year and still observed substantial SCS when restoring desert soils with alfalfa in China.

28

3.8. Reflections on the feasibility of a PVWP carbon offset project

PVWP for restoration of severely degraded lands is a feasible way to simultaneously mitigate desertification and global climate change. The climate change mitigation benefit of the project is comparable to other carbon sequestration projects on the basis of generated mitigation benefit per area (CO2-eq per ha; see Paper I) and it would be the preferable mitigation option per PV system if the excess electricity can be delivered to the grid (see Figure 2). It would, however, also be interesting to investigate the transaction cost of a PVWP carbon offset project. The combination of two fundamentally different mitigation strategies, replacing grid electricity and SCS, may increase the transaction cost. The project is essentially two projects in one and two baselines will have to be constructed. This may therefore add to the cost of implementing the project. The uncertainty of SCS may lower the attractiveness of PVWP projects. If the EF is higher than in Qinghai province or if the SCS potential is significantly lower than estimated it may be more beneficial to simply generate electricity and feed it into the grid to generate carbon offsets. It should, however, be noted that the EF in China is quite high and implementation of a PVWP project may be even more competitive compared to grid electricity substitution in other countries with less fossil fuel dependent electricity sectors. The feasibility is also limited by water constraints, something related to nexus issues that will be discussed in section 4 of the thesis.

29

4. Can water use for carbon sequestration be justified?

“Multiple co-benefits [of dryland restoration], including (…) biodiversity conservation and carbon sequestration, (…) not only helps link the three Rio Conventions † but is an ‘economic way’ to do [climate change] mitigation in drylands.”

- Xie Zhenhua, Vice Chairman, National Development and Reform Commission, China (IISD, 2014).

In this section a model to deal with environmental benefits and concerns of PVWP systems for irrigation of grassland and farmlands in China is introduced. The model is inspired by the water-food-energy-climate nexus approach that is increasingly being applied to areas where water is interlinked with other issues like energy, food production and climate change mitigation (for an excellent overview of the nexus see Hoff, 2011). To answer the question in the title to this section, there are several indicators that could be important to consider. Water use efficiency, water footprint, evaporation recycling ratio, water productivity, and the economic productivity of water are introduced and discussed here.

4.1. Water use efficiency

Chapin et al. (2011) defines water use efficiency (WUE) of photosynthesis as carbon gain per unit of water lost. Xiao et al. (2013) compared the WUE (g C per kg H2O) of different biomes in China to optimize the use of water for carbon sequestration. They concluded that coastal wetlands and forests in South China sequestered much more carbon per unit water than grasslands. The recommendation is

† The three conventions are UNFCCC, Convention on Biological Diversity (CBD), and United Nations

Convention to Combat Desertification (UNCCD), all three are from the Rio summit in 1992.

30

that carbon sequestration measures should focus on forest plantations in southern China since “heat and water are ideal for maintaining high productivity, and this strategy is especially important because efforts to increase carbon sequestration in areas of limited water may inadvertently contribute to the ongoing water crisis in northern China.” (Xiao et al., 2013) Since all plants transpire water vapour, an increase in vegetation will always consume water and reduce runoff that could be used downstream of the project location. A desert would therefore increase runoff compared to vegetated land, and using Xiao with colleagues’ reasoning it may be better to promote desertification rather than combat it. This is clearly unreasonable and even if WUE may seem like an appropriate indicator to justify water use, it is important to take into consideration some other aspects of grassland restoration as well.

4.2. Water footprint

The water footprint of a product is defined as the sum of blue, green and grey water footprints over the entire supply chain. The blue water is the water abstraction from ground and surface-water, and green water is the water contained in soil and vegetation. Grey water is added to the water footprint as an indicator of the freshwater pollution caused by the product (Hoekstra et al., 2011). The water footprint has been promoted as a lifecycle analysis tool for water and shows methodological similarities to the carbon footprint or other ecological footprint concepts. The water footprint is calculated by monitoring the applied irrigation water, the total evapotranspiration (ET) by the plant and by calculating the water needed to dilute any chemicals and fertilisers used in growing the crop in question. The water footprint can then be used to explore ‘comparative water advantages’ between different places to see where it would be most suitable to grow a certain crop.

31

4.3. Evaporation recycling ratio

The continental evaporation recycling ratio, ɛc, is defined by van der Ent et al. (2010) as:

ɛ𝑐 =𝐸𝑐

𝐸 (6)

where: Ec = terrestrial evaporation that returns as continental precipitation E = total terrestrial evaporation (that returns as continental and oceanic precipitation)

In certain areas of the world ɛc is close to 1 and almost all the evaporation will return as rainfall over land. If the ground is covered with vegetation this will increase ET from an area and reduce runoff

(Kleidon et al., 2000). ɛc is very low in South East China, meaning that most water evaporated in these areas will end up in the Pacific Ocean.

ɛc in the grassland sites studied by Xiao et al. (2013) is very high,

much higher than ɛc in the coastal regions where they recommended focusing efforts to increase carbon sequestration. A large share of the water evaporated from the grasslands in Qinghai and Tibet provinces in China will fall as precipitation over the rainfed agricultural areas in North East China (Keys et al., 2012). Considering continental evaporation recycling we can see that justifying water use is complicated by the complex global hydrological cycle and the fact that water evaporated in Qinghai and Tibet will fall as rain over East China can strengthen the case for PVWP projects in this area.

32

4.4. Water productivity

Water productivity is the volume of water needed to produce a product (e.g. m3 per tonne). Water productivity is a function of crop variety and climate parameters, but most importantly yield (Rockström et al., 2014). Water productivity increases until the yield reaches a certain threshold (see Figure 4). Higher biomass production means more vegetation that can shade the soil from short-wave radiation and wind, which decreases direct evaporation of soil water (Ryan et al., 2010). With increased yield there is a ‘vapour shift’ from unproductive evaporation to productive transpiration. Equation 7 describes the relationship between water productivity and yield (from Rockström, 2003).

𝑊𝑃𝐸𝑇 = 𝑊𝑃𝑇

(1 − 𝑒𝑏𝑌) (7)

where: WPET = the water productivity, i.e. the volume ET needed to produce a certain yield [m3 per tonne] WPT = the transpiration needed to produce the yield [m3 per tonne] b = a constant that varies with the local climate conditions [tonne-1] Y = the yield [tonne].

33

Figure 4: Impact on water demand with increasing biomass production as a function of yield. b = – 0.3 and WPT = 350 m3 per tonne.

4.5. The economic productivity of water

The economic productivity of water is a dollar value on what each m3 of water can return (Molden et al., 2010). The use of this indicator can be criticised since it generally is higher for industrial and service products than for agricultural products, which demand the use of more water (Xie, 2009: Figure 3.3). This indicator can therefore justify water allocation to certain sectors while totally disregarding the fundamental need for agricultural products. However, it can be interesting to compare products of similar kind or products produced with different methods in the same region with each other. Paper II discusses the economic water productivity and shows that there is a need to include several other benefits to justify restoration with irrigation, since water is scarce and both production of native grasses (Gao et al., 2013) and alfalfa (Campana et al., 2013a) gives a very small ‘return on water invested’. Xie (2009: p. 37) argues that an area scarce in water resources should import water-demanding products such as beef, pork, rice and wheat. While it is true that beef demands

0

500

1000

1500

2000

2500

3000

0 5 10 15 20

WP

ET W

ate

r p

rod

ucti

vity

(m

3 p

er

ton

ne b

iom

ass

)

yield (tonne biomass per ha)

The dynamics of water productivity and yield

34

a lot of water, other aspects such as if the cattle are grain-fed or grass-fed and cultural aspects of the pastoralists living in the area must also be taken into consideration.

4.6. Which indicators are relevant?

Wichelns (2015) argues that the water footprint concept does not help policy makers in determining the allocation of water. The water footprint does not provide information of arable land or water storage capacity, area under irrigation or use of other technologies in agriculture (ibid). Neither does the concept provide information about opportunity costs for water use and the importance of several other aspects override the water footprint rendering the concept inadequate (ibid). Since irrigation increases yields it reduces non-productive vapour flows, i.e. decreases the value of WPET according to Figure 4, and will help utilise the available water more efficiently for biomass production. This has a clear benefit for the livelihood of pastoralist in the area and can help reduce degradation of grasslands. China is a major importer of alfalfa from the U.S.A. (Bean and Wilhelm, 2010) and the water productivity could be compared with producing alfalfa there. However, in line with Wichelns’ (2015) argumentation there is little to be gained from comparing the water footprint of alfalfa between countries, totally disregarding national food security. Water use is not equivalent to CO2 emissions where the accumulated amount in the atmosphere affects global warming. Instead, water use is about spending the annual water resources sensibly every year, with some but very limited storage possibilities to the next year. This would indicate that the WPET function together with a local water assessment and the variations during dry and wet years are important tools to use. However, water issues cannot be studied in isolation and therefore an integrated nexus model is suggested here in section 4.7, and further described in Paper II.

35

4.7. The nexus model – how do we value the use of water?

Early on in my time in the research project, a colleague of mine (Yan et al., 2013: chapter 7) made an ecosystem service evaluation of grassland restoration to obtain an economic value on the land restauration process (see Figure 5). This approach, to put an economic value on ecosystem services, is a growing research field (see e.g. Farley and Costanza, 2010). The afforestation and reforestation projects in the Gold Standard (2014) project portfolio have the highest co-benefits in monetary terms of all carbon offset project categories (including wind, cook stoves, water filters and biogas), showing that the benefits from land based carbon sequestration projects could be large. However, uncertainties in lifecycle assessments only considering CO2 emissions are large and, depending on what methodology is used, different results can be obtained to justify a variety of standpoints (Grönkvist et al., 2003). Going further to include more environmental variables than CO2 emissions in the lifecycle analysis, reasonably, gives even higher uncertainties. Putting a value on ecosystem services can also be very problematic since it can be used to justify the exploitation of natural resources or negligence of environmental problems based on our extremely limited understanding of ecosystem services. It could also contribute to a ‘commodification of nature’, a privatisation of our common goods (Farley and Costanza, 2010). One example of our limited understanding of ecosystem services is the limited knowledge of how vegetation affects the global hydrological cycle. To my knowledge, no lifecycle assessment has ever taken the value of continental evaporation recycling (see section 4.3) into consideration, despite its importance. In China 80% of the precipitation comes from continental evaporation (van der Ent et al., 2010). Furthermore, monetary values are commonly assumed to exist independently of the person conducting the quantification of an ecosystem service (SOU, 2013). However, in difficult decision situations the value of an ecosystem is dependent on how the questions are asked and if a discussion about the values with the decision makers has taken place (ibid). All these aspects make quantification of the value of grassland

36

restoration problematic. The estimated economic values of the ecosystem services are presented in Figure 5.

Figure 5: Estimated economic value (USD per ha per year) of the ecosystem services gained by grassland conservation (from Yan et al., 2013). Contrary to the economic lifecycle analysis, a conceptual model is used in Paper II to analyse and justify the use of water in a water-constrained area. The model is not all-inclusive, and several other co-benefits could be added to it. Instead, the model is an attempt to conceptualise the environmental benefits of a PVWP project. In Paper II, we refrain from putting economic values on the benefits of the project since this tends to direct the discussion on comparing economic costs and benefits, and disregarding other values. Paper II shows that the carbon offset project cannot be justified using the economic productivity of water. The economic productivity (USD per m3) of the products produced by the PVWP system is very low

Org

anic p

rodu

ctio

n

Carbo

n Seq

uestr

atio

n

Nut

rient

cyc

ling

Water

con

serv

atio

n

Soil e

rosio

n co

ntro

l

Air

clea

ning

0

100

200

300

400

500

600

700

$

Ecosystem services

US

D p

er

ha a

nn

uall

y

37

(Olsson et al., 2015: Table 2), hence based only on the economic water productivity the carbon offset project should be rejected. However, it is important to recognise that several of the co-benefits to carbon sequestration generated by the project are hard to value in monetary terms. The justification must instead include other aspects and Paper II introduces a nexus model as an alternative way to analyse and justify water use. The model is similar to a SWOT-analysis (see Table 1; SWOT is short for Strength Weakness Opportunities and Threats) and can be used to identify benefits and concerns with irrigation and other water management options in dryland agriculture. It can provide a conceptual basis to discuss a PVWP project and the impacts it may have on a local scale as well as on a regional to national scale. There are several ways of integrating and pricing ecosystem services in policy making. The main difference between the monetary approach and the qualitative approach is that a tax or another incentive is set to reach the goal of the policy in the latter case. With a monetary approach it is instead up to the market to determine the outcome of the policy.

38

Table 1: Policy decision basis for integrating ecosystem services in

the decision making process (modified from SOU, 2013: p. 206).

Qualitative Quantitative Monetary

Methods

for

description

of values.

Base for

decision

making.

SWOT-analysis.

Historical

mapping.

Identification of

ecosystem services.

Stakeholder

dialogue. Multi-

criteria analysis.

Quantitative

mapping of

ecosystem

services, e.g.

water flows,

pollination.

Multi-criteria

analysis.

Monetary

calculations of

values for

water flows,

pollination etc.

Cost-benefit

analysis.

Policy integration through direct effect on prices.

Tax. Environmental offset with several goals. The tax is set to reach the goal of the policy, not to correspond to the external benefit.

Tax. Environmental offset with a quantitative goal. The tax is set to reach the goal of the policy, not to correspond to the external benefit.

Tax. Environmental offset with a monetary value which shall correspond to the external benefit.

When describing the benefits and concerns, different stakeholders that can be deemed important to the project implementation will arise. Figure 6 illustrates how the model can be implemented to aid understanding of the multiple co-benefits to the project, something that will be important in the stakeholder dialogue. Different components in the model are interconnected with each other in the nexus.

39

Figure 6: (a) A nexus model with two complementary system boundaries considering water availability, carbon sequestration and sustainable development on project scale (grey boundary) and water, food and energy security on a suitable, larger, nexus scale (blue boundary). (b) The project scale (grey boundary) enlarged.

4.8. Reflections on carbon sequestration, water and co-benefits

Justifying water use to sequester carbon is not straightforward considering the multiple co-benefits associated with carbon sequestration projects. There are a multitude of indicators for water

40

use and some look very appealing since they produce a quantitative value that can be compared with other values to draw a conclusion. The water footprint can be used to justify import of goods from other locations where the water footprint is lower, but this indicator disregards all other considerations such as food security and carbon sequestration. The water use efficiency has been used to criticise carbon sequestration in the Chinese grasslands (Xiao et al., 2013), but the criticism is out of context since it focuses too narrowly. These simplified policy aids are appealing since they provide a single number comparable to other single numbers, but it can say very little about the actual benefit of carbon sequestration in drylands. In the same simplified spirit, the recommendation by Xiao et al. (2013) to focus carbon sequestration efforts to the south of China, and using water use efficiency as the only indicator upon which to base this argument, is somewhat problematic. China is depending on import of alfalfa from the U.S.A. and providing opportunities for domestic alfalfa production will increase national economic independency and food security. Therefore, the water footprint is misleading; it is too simple and does not consider the complex decision situation. Several other co-benefits must be taken into consideration and according to the results in Paper II, carbon sequestration in the Inner Mongolia and the Qinghai province can be one feasible way to provide multiple co-benefits to mitigation of climate change. The nexus model can also be used as a risk assessment tool, or a SWOT-analysis. Although the primary goal of a carbon sequestration project is to create carbon credits in a sustainable way, and not to gain incomes from pricing ecosystem service, it is a necessity to assess the impact on the ecosystem in which the project operates. Failure to address ecological concerns of carbon sequestration will endanger the permanence of the sequestered carbon.

41

5. Policy aspects of PVWP project implementation

“Leakage is defined as the net change of anthropogenic emissions by sources of greenhouse gases which occurs outside the project boundary, and which is measurable and attributable to the CDM project activity. The word measurable is in this case problematic, as many of the effects that, potentially, could generate significant leakage effects, are not measurable.”

- Stefan Grönkvist, researcher and systems thinker, KTH (Grönkvist, 2005).

Mitigation of climate change is taking place on several levels in society. On the global political level, countries are meeting within the United Nations to discuss how policy should be implemented to best curb accumulation of GHGs in the atmosphere and this political level is widely considered as the most important arena for global mitigation. There is a trade-off between measurable and verifiable mitigation actions on one hand and implementation of mitigation projects on the other. Generally speaking, the more rigorous reporting demands are on mitigation projects, the less likely they are to be implemented. In this section I will argue that mitigation efforts in developing countries have been slowed due to the exclusion of several important carbon sequestration activities from the CDM, which so far is the most important mechanism within the UNFCCC to involve developing countries in mitigation of climate change. This section also assesses the risk and economic feasibility of the PVWP project.

5.1. Leap frogging into a clean future

Winkler and Thorne (2002) asked “should development be allowed to get dirty before it qualifies to become clean?”

42

and they argued for promoting leap frogging of development into a ‘clean’ state without transitioning through a ‘dirty’ state where CO2 emissions increase as a function of economic development. The pace with which we are combusting the remaining carbon budget of a cumulative value of about 1000 billion tonne carbon (IPCC, 2013: Chapter 12), releasing GHGs to the atmosphere and approaching dangerous warming of the planet calls for a mechanism to allow non-Annex I countries to develop without using fossil fuels. The mechanism thus far has been the CDM. CDM has a leading role in determining what rules and regulations should govern carbon offsets also on the voluntary markets.

5.2. Policy mechanisms for engaging developing countries in mitigating climate change

CDM’s fate under the new Paris climate agreement is unknown. However, the idea that mitigation of climate change can be made more cost-effective through emission trading and flexibility is likely here to stay, and discussions about how to simplify CDM even further is on-going (UNFCCC, 2015). It is therefore likely that grassland restoration projects will be a significant part of future mitigation strategies, especially in developing countries. A step towards including all Parties to the UNFCCC in mitigating climate change was taken during the climate meeting in Bali. The Bali Action Plan paragraph 1 (b) (ii) considers “Nationally appropriate mitigation actions by developing country Parties in the context of sustainable development, supported and enabled by technology, financing and capacity-building, in a measurable, reportable and verifiable manner” (UNFCCC, 2008) While the first time NAMAs was mentioned was in Bali, the COP 16 meeting in Cancun, Mexico, agreed that

43

“developing country Parties will take nationally appropriate mitigation actions in the context of sustainable development, supported and enabled by technology, financing and capacity-building, aimed at achieving a deviation in emissions relative to ‘business as usual’ emissions in 2020“ (UNFCCC, 2010: paragraph 48). It was also decided: “to set up a registry to record nationally appropriate mitigation actions seeking international support and to facilitate matching of finance, technology and capacity-building support for these actions” (ibid: paragraph 53). Cancun was also an important step to engage developing countries in enhanced reporting and it was decided that the NAMAs receiving international support should be subject to international measurement, reporting and verification (MRV; ibid: paragraph 61). The Green Climate Fund (GCF) was set up at the same meeting. The fund is going to be handling the multilateral funding for adaptation and mitigation of climate change with a goal of USD 100 billion per year by 2020 (ibid: paragraph 98). The GCF has reached its first threshold of USD 10 billion in 2014 (GCF, 2014) showing that the fund has the possibility to act as the financial mechanism in line with article 11 of the Convention (UNFCCC, 1992).

5.3. Uncertainties of carbon offsets and NAMAs

The international funding of NAMAs will be conditioned by MRV in much the same way as the CDM has been. The difference between CDM and NAMAs is the scale of the mechanism and the transformational effect that NAMAs should have (UNFCCC, 2014c). Uncertainties of the emission reductions achieved by CDM projects are large, mostly due to rebound effects. The rebound effect is a phenomenon occurring wherever energy efficiency or fossil fuel substitution is used to claim emission reductions. If a new product can perform a service, using less energy than an old product, energy

44

use will decrease – if everything else in the economy stays the same. However, change will usually occur. Because the equipment has become more energy efficient, the cost per unit of services of the equipment falls. A price decrease could lead to increased consumption and part of the energy efficiency is used instead of conserved. The rebound effect can be defined as this lost part of the energy conservation (based on Berkhout et al., 2000). The rebound effect can lead to a number of accounting problems when trying to quantify GHG emission reductions. Several CDM projects are energy efficiency related; they claim to reduce CO2 emissions by increasing energy efficiency relative to the baseline energy efficiency. A classic example is the efficiency in the conversion process from chemical energy in a fuel to indoor thermal energy. If a house is well insulated, less heat is lost as waste heat and less fuel must be burnt to generate a given indoor temperature. However, the whole benefit of the efficiency improvement is only obtained if the indoor temperature does not change and this may not always be the

case. If the desired indoor temperature is 20°C and the temperature

before insulation is 17°C, then some of the efficiency will probably be used to achieve the desired indoor temperature, hence there is a rebound effect. Even though this will not always happen, it is easy to understand that rebound effects are larger in developing countries where there is a suppressed demand due to economic reasons or limited availability of resources such as biomass for heating. CDM projects could therefore suffer from large rebound effects and policy makers must be aware of this, although it is irrational to let it paralyse the implementation of mitigation actions. Paper III illustrates how uncertainties due to rebound effects may not be safely ignored, and thus they should impact the way CDM and other mitigation efforts, like NAMAs, are shaped. Energy efficiency may bring economic growth and many CDM projects are likely to stimulate development. If CDM stimulates a country’s development in certain sectors of the economy, then development in the LULUCF sector is under-stimulated due to the restrictions put on it. Including additional LULUCF activities in the CDM may influence NAMAs

45

and engage several of the least developed countries with large potentials to mitigate climate change using LULUCF activities. It has indeed been claimed that the exclusion of LULUCF activities is one reason for the small amount of CDM projects in Africa (Whitman and Lehmann, 2009). Instead, CDM has been focused on the fast-growing economies such as China, the country that creates almost 60 % of the CERs (UNEP, 2014). Knowledge of the uncertainties that are part of the CDM would increase awareness and could perhaps decrease the amount of unsound CDM projects. One could argue that the need for a mechanism to engage developing countries in the mitigation process is greater than the need of absolute certainty of emission reduction from this mechanism. Striving to achieve absolute certainty can paralyse policy makers and shift focus from practical solutions to deal with an important problem. However, it is at the same time important to acknowledge the uncertainties and to improve the methods to deal with them. The future use of CDM or a similar mechanism should involve recognition of the immeasurable uncertainties in the MRV process, not to decapitate CDM but to gain awareness of the uncertainties. The fact that some uncertainties of leakage – like rebound effects – are excluded is due to that they do not fulfil the criteria of being measurable and attributable to the project activity (more about leakage in section 3.4 Leakage of this thesis). The difference between rebound effects and uncertainties in the LULUCF sector is that the latter is measurable, but to a high cost. But the uncertainties may be quantitatively similar (see Paper III) and it seems inconsistent to treat them differently. Hence, the uncertainty of LULUCF activities is therefore not a sufficient argument to exclude them from the CDM. If a broad range of LULUCF activities will be included in a future CDM, methodologies may be developed to deal with uncertainties and every host country’s cost-benefit analysis will determine if the CDM project is feasible or not.

46

5.4. Non-permanence risk assessment

The inherent difference between carbon sequestration and emission reduction is the permanence of the carbon sink. A sink is only a sink until the carbon in the sink is emitted to the atmosphere, emission reductions last forever. That is at least the attitude in the CDM. However, to be able to compare carbon sequestration projects there is a need to compare emissions of GHGs with sequestration of CO2. Furthermore, it also requires addressing the risks of reversal of the carbon sink. The VCS LULUCF projects are required to assess the risk of non-permanence (VCS, 2012a). The risk assessment is made up by three categories, namely internal, external and natural risk. Each category is subdivided and the subcategories are added up to a total value for the three risk assessments (see Table 2). This procedure makes it possible to balance a high risk in one category with a low risk in another, and thus a low financial viability can be compensated for by, for example, a low opportunity cost for the project area. Opportunity cost can be shown to be low if the land is degraded and the value of alternative activities on the land is low, and it can be further enhanced by legally binding long-term commitments to restoration of land. The subcategory project management is related to the suitability of the species used to achieve carbon sequestration and the skills of the project team. The project longevity has to be at least 30 years, or else the project fails the risk assessment and is not eligible for carbon credits under the VCS (VCS, 2012a: p. 10). Legally binding project longevity of 60 years gives a score of 0 in the risk assessment subcategory. The external risks further include community engagement, which can be mitigated by generating positive local impacts, and political risk. The natural risk to the project further includes risk of severe weather events, pest outbreaks, fire risk etc. that can reverse the carbon sink. The risk assessment of a hypothetical PVWP project is illustrated in Table 2.

47

Table 2: Non-permanence risk assessment as suggested by VCS (2012a)

Risk Score

- Project management

- Financial viability

- Opportunity cost

- Project longevity (at least 30 years)

Internal risk =

-4

6

-4

2

0

- Land and resource tenure

- Community engagement

- Political risk

External risks =

7

-5

2

4

Natural risks = 2

Sum of total risk = 6

According to the VCS tool to assess risk (VCS, 2012a), the minimum risk rating is 10, even if the risk assessment shows a lower value. An overall risk rating of over 60 is unacceptably high and the project fails the risk assessment. A risk rating of 6, which therefore takes the value 10, as for the hypothetical PVWP project illustrated in Table 2, converts to a 10 % risk buffer of the claimed LULUCF sink against reversal. A risk rating of 20 would mean to that 20 % of the carbon sink would have to be offset to the buffer. That 10 % may be a realistic buffer is shown by the VCS project the Kenya Agricultural Carbon Project, implemented by the non-governmental organisation Vi Agroforestry, that has to deposit 10 % of their credits to the VCS buffer fund to insure against future reversal (Tennigkeit et al., 2013).

5.5. The permanence period

Is it possible to compare emission reduction and carbon sequestration? There is a real difficulty in comparing emission reduction and carbon sequestration. Emission reduction can be measured and accounted for at the time when they take place,

48

whereas a carbon sink has to be monitored for a certain time, called the ‘permanence period’ (UNFCCC, 2014b). How to calculate the permanence period depends on the approach. By looking at the radiative forcing effect of CO2 over time, Moura-Costa and Wilson (2000) concluded that carbon sequestered for 55 years is equivalent to permanent sequestration. This approach is called ‘tonne-year’, since sequestration of carbon is related to the radiative forcing effect of one tonne of CO2. The tonne-year factor, or equivalence factor, is then 0.0182 (1/55) tonne of CO2-eq emission reduction, meaning that sequestration of one tonne CO2 for one year has the same effect as emission reduction of 0.0182 tonne CO2. The tonne-year approach is very appealing to use due to its simplicity. Furthermore, it makes it possible to reward carbon sequestration over shorter time periods, without the demand and risk of multi decadal project commitments. The approach is complicated by the fact that the IPCC special report on LULUCF (IPCC, 2000: Chapter 2.3.6.3) mentions permanence periods from 42 to 150 years, depending on which approach that is used for the calculation. Global warming potential for different GHGs are considered over a policy-relevant period of 100 years (UNFCCC, 2012a: Decision 4, paragraph 5). Thus, another perspective may therefore be that the sink is only permanent after 100 years, since this timeframe is the commonly agreed policy period. Whatever the permanence period is, the project will require a sound business idea that can sustain the incentive to keep the carbon sink permanent. Grass production that is profitable also in the future is one such idea and this is utilised in PVWP projects to give incentives to keep the carbon in the ground. But the future of water scarcity and the opportunity cost of the land are both very uncertain in the Chinese grassland area and could threaten the permanence of the carbon sink.

5.6. Economic assessment

I will here assess the economic feasibility of a case project in Qinghai province, China, described by Campana et al. (2013a) both as a

49

comparison with data from CDM PV projects and as an additionality test. To be economically feasible a PVWP project requires alfalfa production of around 8 tonne of dry matter (DM) per ha and year (ibid), which is a high value for degraded land. A more realistic yield may be around 4 tonne DM per ha annually (ADB, 2010). The most sensitive variable to the feasibility of the system is the income from grass production (Campana et al., 2013a). Including the cost of well construction and the investment cost of the irrigation system makes the project unfeasible at 3.8 tonne DM per ha, when only considering incomes from alfalfa production (ibid: Figure 7). However, including sales of surplus electricity from the PV system and income from carbon offsets at USD 20 per tonne CO2-eq could significantly improve the profitability of the system and shows economic feasibility up to an internal rate of return (IRR) of 10.9 %. The same project without income from carbon offsets would require an IRR lower than 8.8 %. Figure 7 compares the required IRR for a project with and without CERs at USD 20 per tonne CO2-eq.

50

Figure 7: The net present value (NPV) per ha of project area for different IRR. The project produces 3.8 tonne DM alfalfa per annum with a price of USD 207 per tonne DM. The income to the project includes sales of excess electricity and feed-in tariffs for PV (cf. Campana (2015) for details on the economic feasibility). In CDM, additionality can be demonstrated using a benchmark analysis, thus comparing the IRR from a project with CERs, with the absence of the additional income from CERs. PV projects in China often use the benchmark 8 % IRR. However, several other barriers to implementation of the project other than economic barriers can exist (see section 3.3 Additionality of this thesis). For the benchmark IRR of 8 %, the project NPV without CERs is USD 530 per ha and USD 2000 including CERs at a carbon price of USD 20 (Figure 7). The profitability without the additional income from the CERs makes the project fail the additionality test and additionality is required to be demonstrated in some other way. Table 3 presents the IRR from five

51

PV projects from the Qinghai province in China to compare the grid connected PV project with the PVWP project. Table 3: IRR for five CDM projects employing grid connected PV in Qinghai province, China.

Project reference number

IRR without CERs (%)

IRR including CERs (%)

8051 6.28 7.54

7962 6.05 No data

7991 5.39 6.31

8013 6.53 8.05

5952 6.42 7.29

The IRR from the PVWP case project is higher than the IRR from the CDM PV projects, and if additionality can be proven the feasibility of PVWP projects as carbon offset projects seems to be good.

5.7. Water policy

As shown in Paper II, PVWP systems have the potential to improve water productivity (WPET, see Figure 4). However, productive, or efficient, use of water and energy is very different concepts. Inefficient use of energy will result in irreversible heat-losses, whereas inefficient water-use will lead to increased ET or increased runoff. We do not benefit from ‘saving’ water, the same way we benefit from ‘saving’ energy. Assuming we do not use fossil water from deep aquifers, water management is about spending the annual water budget wisely and being aware of trade-offs between downstream and upstream water users. In addition to this, continental evaporation recycling is an important concept to justify water use for carbon sequestration, since water that is evaporated in West China will fall as rain over East China, and thus the geographical location of the PVWP project is very important.

52

Allocation of water among sectors is usually a political decision and not governed by market mechanisms (Hellergers and Perry, 2006), and energy use is therefore quite different to water use. Water use is a matter of choosing between activities and the choice is not necessarily based on the economic value of the water-using activity. Paper II puts forward a host of good reasons to use water for carbon sequestration. Water and climate change policy are interconnected and may therefore benefit from being developed jointly, something that becomes clear in carbon sequestration projects. Furthermore, the water and climate change issues are linked to food and energy security, demanding a holistic multidisciplinary perspective from policy makers on carbon sequestration in drylands.

5.8. Reflections on policy issues and the economy of PVWP projects

Section 3 of this thesis introduced three alternative carbon offset projects and compared their feasibility (see Figure 2). PVWP systems used for restoration of degraded lands can produce 148 CERs per ha degraded land during 20 years, according to Paper I. The crediting period for emission reduction projects in the CDM is seven years, three times renewable. This would give an operating time for renewable energy projects of 21 years. Using the same system, in the same location, for 21 years would generate 92 CERs if electricity was fed directly to the grid instead. Substituting diesel for an equivalent pumping system during 21 years would yield 39 CERs. There is thus an opportunity to increase the mitigation benefit of PV systems in the CDM if policy would change and grazing land management was to be included as an eligible activity. The economic feasibility of the PVWP system is very uncertain at low levels of grass production, something that could be expected on degraded soils. However, the extra income from electricity generation and feed-in tariffs could make the project feasible also on degraded lands. Given the many uncertainties in the economic feasibility analysis, CERs can clearly play a vital role in attracting investment to the projects. If the project is financially attractive without CERs, it

53

may also be implemented without the additional incentive that CERs offer. Furthermore, it is important to recognise the importance of the co-benefits that PVWP projects bring about compared to grid connected PV projects, despite the lack of economic gains for these additional ecosystem services. The economic feasibility can appeal to private investors and the CERs can help to make investments in otherwise economically unfeasible project areas possible. In addition, the ecosystem services connected to the PVWP projects can appeal to local and national governments. This shows that a broad range of funding bodies can be possible targets for project implementation. The increase of grass production due to the PVWP project will provide more forage to the grazing animals and could potentially lead to an increase in the amount of animals instead of conservation of grassland areas. It is important to keep the conservational scope of the project if the mitigation benefit should be realised. On a national level, efficiency improvements may be supplemented with conservation or restoration policies to ensure that efficiency improvements have the intended effect. The efficiency improvements on local scale, e.g. in a CDM project, must be combined with policies on regional or national scale to make sure that the improvements actually achieves climate change mitigation.

54

6. Discussion

6.1. The climate change mitigation potential

Using PVWP systems to combat degradation of grasslands seems to be a feasible approach that simultaneously can mitigate climate change and adapt grassland management to climate change. Paper I developed a methodology for such a project. The PVWP project achieves climate change mitigation in two ways: it sequesters carbon in the soil during the restoration process, and it makes use of the excess electricity from the PV system to supply grid electricity. The transaction cost of the project has not been investigated in this thesis, but this dual project nature could add costs to the implementation of PVWP projects. However, according to a decision from CDM EB, there is a positive list of automatically additional CDM projects, including grid connected PV systems up to 15 MW (UNFCCC, 2011) making the transaction cost for these projects low. Furthermore, if additional CERs can be produced in an area with this approach compared to using the PV system to solely produce grid electricity, this may justify the increased transaction cost. The economic feasibility analysis performed in this thesis shows that the internal rate of return (IRR) could be higher for a PVWP project than for a grid electricity PV project in the CDM. Although the analysis is very site-specific, in the investigated case it is necessary to utilise the excess electricity produced by the PV solar modules to make the PVWP system economically feasible when restoring degraded grasslands. The novel baseline scenario where severely degraded grasslands remain in a state with low levels of SOC is investigated and compared to other baselines. Compared to the alternative projects (A1 and A3 in Figure 2), the carbon offset project proposed in Paper I of this thesis (A2* in Figure 2) generates much higher climate change mitigation potential. The proposed project sequesters carbon in the soil and monitoring this mitigation benefit is fundamentally different and more complex compared to monitoring the alternative project outcomes. A methodology developed by BioCarbon Fund and approved by VCS (VCS, 2011) with monitoring of biomass input to the soil and modelling using RothC model can be

55

a feasible and cost-effective way of implementation (World Bank, 2013). However, this remains to be studied. It is interesting to note that replacing diesel powered irrigation systems with PVWP systems will not be a viable carbon offset project at present in the investigated area. A PV system will almost always generate more CERs if it is used to generate electricity for the grid. Instead, the opportunities in using PVWP as a carbon offset project lies in restoration of severely degraded grasslands. The mitigation benefit of using PV systems to run the irrigation system instead of feeding electricity to the grid is much higher due to the large soil carbon sequestration potential in degraded soils. The ultimate feasibility of PVWP systems as carbon offset projects depends on the electrical grid EF. If soil carbon sequestration possibility is low and the EF is high, then grid connected PV projects yields higher mitigation benefits. When developing countries increase their use of renewable energy sources, the grid EF will likely decrease and this would increase the attractiveness of PVWP systems compared with grid connected PV projects.

6.2. Water limitations

While PVWP carbon offset projects have potentially high co-benefits, water scarcity is a real concern and must be addressed. A range of indicators to deal with water-use are analysed in this thesis. The water use efficiency and the water footprint seem to give little meaningful information to the assessment of PVWP systems as carbon offset projects. Using a single assessment to find appropriate locations for carbon sequestration, like the water use efficiency proposed by Xiao et al. (2013), has limited use. Although the two concepts can provide information to the bigger picture, it is important to view the water allocation decision process from a local and national perspective as well. China is dependent on import of alfalfa from the U.S.A. and providing opportunities for domestic alfalfa production will increase the national economic independency and food security, despite the value of the water footprint. While the economic productivity of

56

water (USD per m3) is low for the proposed carbon offset project, PVWP projects have a multitude of co-benefits and can contribute to several positive outcomes other than high economic output. To reflect the co-benefits, a nexus approach is suggested in this thesis (and discussed at length in Paper II), including several assessments to illustrate the co-benefits of the carbon offset project. The suggestion of how this nexus model could look is given in Paper II and includes climate change mitigation benefits, blue water availability, energy security, downstream water impact, food security and continental evaporation recycling ratio.

6.3. The complexity of project implementation

It is interesting to look at ways that policy makers can deal with the trade-offs in climate change policy implementation. There is no silver bullet answer to this question. Grönkvist (2005: p. 58) suggests that using “a pragmatic, experience-based method utilizing information from various kinds of sources, including models, (…) is probably a much more adequate approach for finding efficient ways of how to tackle the GHG problem than the use of an over-simplified model.” At the same time there is a need to simplify the methodologies for carbon offset projects since implementation costs could be very high and prevent the feasibility of projects. In Paper III it is argued that uncertainties in several categories of the CDM are of the same magnitude as the uncertainties of LULUCF activities. This provides a strong case for LULUCF in CDM, if certain uncertainties can be acceptable. Including additional LULUCF activities in the CDM may also engage the least developed countries in mitigation of climate change, given the large potential in restoration of degraded lands in the agricultural sector. Thus far, however, countries with high electricity consumption and industrial development have been the major hosts to CDM projects, severely limiting the scope for engaging the least developed countries of the world in mitigating climate

57

change. Furthermore complexity is added to the PVWP project idea as there is a multitude of site-specific co-benefits to take into consideration. Although it will strengthen the case for the project, monitoring of co-benefits and water use, in addition to monitoring the carbon sink, would add extra costs and would have to involve stakeholders benefiting from the project and potentially suffering from reduced downstream water supply. This is a challenge for all LULUCF projects. To develop a practical approach to the co-benefits and to cater for the fact that monetization of ecosystem services is a very costly and inexact way of valuing these co-benefits, it is proposed in this thesis that a qualitative approach to policy development should be taken. Instead of letting the market decide if grasslands are valuable enough to be restored, a nexus approach has been proposed to show how the grasslands are important for climate, water, food and energy security. While this nexus-thinking could form a foundation for a national or regional policy that will increase the restoration of degraded grasslands, it is also a way to justify water use for carbon sequestration in drylands, making carbon offset projects possible in this area of the world.

58

7. Conclusions

A novel baseline scenario is proposed in this thesis where emission reduction is combined with carbon sequestration to assess the full potential of PV water pumping (PVWP) carbon offset projects. This novel baseline shows that the potential for PVWP projects as carbon offset projects is higher for restoration of degraded land than for projects where diesel powered irrigation is replaced with PV powered systems. In most cases, using the PV system to generate electricity for the grid is much more beneficial than to replace diesel due to the inability of the pumping system to utilise the total amount of electricity produced by the PV system. The alternative use of the PV system to mitigate climate change by instead supplying electricity to the grid, substituting fossil fuel power production, is a function of the grid emission factor (EF). If power production is associated with low fossil fuel emissions, PVWP will compete better as a carbon offset project with grid electricity. This means that as developing countries install more renewable power the attractiveness of the PVWP system as a carbon offset project will increase.

Water for carbon sequestration is a controversial subject, but the irrigation project can achieve a vapour shift from unproductive evaporation to productive transpiration, which increases utilisation of rainfall for biomass production. Producing more biomass per water unit, it is possible to build healthy grasslands that could bring about several ecosystem services and fulfil several sustainable development criteria. These environmental co-benefits should be compared with potential negative impacts on downstream water users. Using the continental evaporation recycling ratio can further support PVWP project implementation, especially in the Qinghai and Tibet provinces in China.

The economic feasibility analysis shows that the income from sales of excess electricity is crucial for the project and that the

59

additional income from certified emission reductions (CERs) has a significant impact on the project’s feasibility.

There are several potential funders of PVWP projects. The economic feasibility can appeal to private investors and the CERs can help to make investments in otherwise economically unfeasible project areas possible. Furthermore, the ecosystem services connected to the PVWP projects can appeal to local and national governments.

The thesis also argues for including additional LULUCF activities, such as grazing land management, in the CDM. The discussion provides new insights to the uncertainties in different CDM categories for the ongoing debate of whether to include additional LULUCF activities in the CDM or not. Until then, the only possibility to explore opportunities with PVWP systems as carbon offset projects is on the voluntary carbon markets.

60

8. Future studies

Scope for future research is given here:

Looking at implementation of a carbon offset project in practice, with transaction costs, local assessment of carbon inputs and modelling with RothC for a specific case.

Grasslands are not purely vegetated by grasses, but include small shrubs and trees. Integrating trees and shrubs in the grassland to create wind breaks, to slow down water flows and increase moisture infiltration could potentially sequester more carbon and create a multifunctional agroecosystem (Ryan et al., 2010).

Implement and evaluate the nexus model at a stakeholder meeting and try to understand the feedbacks and interconnections for a real case. Furthermore, it could be interesting to investigating the use of facilitating tools like the one developed by Stockholm Environment Institute (SEI) called Water Evaluation and Planning (WEAP) (Karlberg, 2014).

Monitor the progress of CDM and NAMAs and investigate how capacity building under CDM influences NAMAs in the development process.

61

References

ADB (Asian Development Bank), 2010. Qinghai pasture conservation using solar photovoltaic (PV)-driven irrigation. ADB RSC-C91300 (PRC). Akiyama T, Kawamura K, 2007. Grassland degradation in China, Methods of monitoring, management and restoration. Grassl Sci 53(1), 1–17. doi: 10.1111/j.1744-697X.2007.00073.x Bean R, Wilhelm J, 2010. U.S. alfalfa exports to China continue rapid growth. GAIN report number: CH10054. USDA Foreign Agricultural Service. Berkhout PHG, Muskens JC, Velthuijsen JW, 2000. Defining the rebound effect. Energ Policy 28(6-7), 425–432. doi: 10.1016/S0301-4215(00)00022-7 Boyd E, Corbera E, Estrada M, 2008. UNFCCC negotiations (pre-Kyoto to COP-9): what the process says about the politics of CDM-sinks. Int Environ Agreem 8(2), 95-112. doi: 10.1007/s10784-008-9070-x Campana PE, 2015. PV water pumping systems for agricultural applications. Available from: http://www.diva-portal.org/smash/get/diva2:792682/FULLTEXT02, accessed 12 April 2016. Campana PE, 2013. PV water pumping systems for grassland and farmland conservation. Available from: http://www.diva-portal.org/smash/get/diva2:666134/FULLTEXT02.pdf, accessed 14 October 2014. Campana PE, Olsson A, Li HL, Yan JY, 2013a. Economic analysis of photovoltaic water pumping irrigation systems. In proceedings of the International Conference on Applied Energy 2013, July 1-4, Pretoria, South Africa. Campana PE, Li HL, Yan JY, 2013b. Dynamic modelling of a PV pumping system with special consideration on water demand. Appl Energy 12, 635–45. Campana PE, 2015. PV water pumping systems for agricultural applications. Mälardalen University Press Dissertations No. 175. Available at: http://mdh.diva-portal.org/smash/get/diva2:792682/FULLTEXT02.pdf, accessed 24 June 2015. Carrasco LR, Larrosa C, Milner-Gulland EJ, Edwards DP, 2014. A double-edged sword for tropical forests. Sci 346(6205), 38-40. doi: 10.1126/science.1256685 CDM EB (Clean Development Mechanism Executive Board), 2013. Annual Report 2013. Available from:

62

http://unfccc.int/resource/docs/publications/pub_cdm_eb_annualreport_2013.pdf, accessed 10 October 2014. CDM Rulebook, 2014 [Internet]. Baker & McKenzie. Swedish Energy Agency, Australian Government Department of Climate Change. Available from: http://cdmrulebook.org/166, accessed 19 October 2014. Chapin FS III, Matson PA, Vitousek P, 2011. Principles of Terrestrial Ecosystem Ecology 2nd edition. Springer, New York Dordrecht Heidelberg London. doi: 10.1007/978-1-4419-9504-9 China’s INDC, 2015. Enhanced Actions On Climate Change: China’s Intended Nationally Determined Contributions (Unofficial translation). Available from: http://www4.unfccc.int/submissions/INDC/Published%20Documents/China/1/China's%20INDC%20-%20on%2030%20June%202015.pdf, accessed 12 April 2016.

Chinese DNA (Designated National Authority), 2010. 中国区域电网基准线排放

因

子[Chinese: 2010 Baseline Emission Factors for Regional Power Grids in China]. Available from: http:// cdm.ccchina.gov.cn/WebSite/CDM/UpFile/File2552.pdf, accessed 3 February 2015. Conant RT, Paustian K, Elliott ET, 2001. Grassland management and conversion into grassland: Effects on soil carbon. Ecol Appl 11(2), 343–355. Conant RT, 2012. Grassland Soil Organic Carbon Stocks: Status, Opportunities, Vulnerability. In: Lal R, Lorenz K, Hüttl RF, Schneider BU, von Braun J (eds.). Recarbonization of the Biosphere: Ecosystems 275 and the Global Carbon Cycle. Springer, Dordrecht Heidelberg New York London. pp. 275-302. doi: 10.1007/978-94-007-4159-1_13 Dai A, 2011. Drought under global warming: a review. Wiley Interdisciplinary Reviews: Clim Chang 2(1), 45-65. doi: 10.1002/wcc.81 Dile YT, Karlberg L, Temesgen M, Rockström J, 2013. The role of water harvesting to achieve sustainable agricultural intensification and resilience against water related shocks in sub-Saharan Africa. Agric Ecosyst Environ 181, 69–79. doi: 10.1016/j.agee.2013.09.014 Erickson P, Lazarus M, Spalding-Fecher R, 2014. Net climate change mitigation of the Clean Development Mechanism. Energy Policy 72, 146–154. doi: 10.1016/j.enpol.2014.04.038

63

Fan MM, Li YB, Li WJ, 2015. Solving one problem by creating a bigger one: The consequences of ecological resettlement for grassland restoration and poverty alleviation in Northwestern China. Land Use Policy 42, 124–130. doi:10.1016/j.landusepol.2014.07.011 Farley J, Costanza R, 2010. Payments for ecosystem services: From local to global. Ecol Econ 69(11), 2060–2068. doi:10.1016/j.ecolecon.2010.06.010 Gao XR, Liu JH, Zhang J, Yan JY, Bao SJ, Xua H, Qin T, 2013. Feasibility evaluation of solar photovoltaic pumping irrigation system based on analysis of dynamic variation of groundwater table. Appl Energy 105, 182–93. doi: 10.1016/j.apenergy.2012.11.074 GCF (Green Climate Fund), 2014. Green Climate Fund hits USD 10 billion threshold. Press reslease. Available from: http://news.gcfund.org/wp-content/uploads/2014/12/release_GCF_2014_12_05_australia_belgium_pledge.pdf, accessed 14 January 2015. Gold Standard, 2014. The real value of robust climate action – Impact investment far greater than previously understood – A net balance report for the gold standard foundation. Available at: http://www.goldstandard.org/wp-content/uploads/2014/05/GoldStandard_ImpactInvestment.pdf, accessed 4 February 2014. Gordon LJ, Enfors EI, 2008. Land Degradation, Ecosystem Services and Resilience of Smallholder Farmers in Makanya Catchment, Tanzania. In: Bossio D, Geheb K (eds.). Conserving Land, Protecting Water. CAB International, Reading, UK. pp. 31-50. Grönkvist S, 2005. All CO2 molecules are equal, but some CO2 molecules are more equal than others. Universitetsservice US AB, Stockholm, Sweden. See also: http://www.diva-portal.org/smash/get/diva2:12657/FULLTEXT01.pdf, accessed 8 January 2014. Guan DB, Hubacek K, 2007. Assessment of regional trade and virtual water flows in China. Ecol Econ 61(1), 159–170. doi:10.1016/j.ecolecon.2006.02.022 Hellegers PJGJ, Perry CJ, 2006. Can Irrigation Water Use Be Guided by Market Forces? Theory and Practice. Int J Water Resour Dev 22(1), 79-86. doi: 10.1080/07900620500405643 Henderson BB, Gerber PJ, Hilinski TE, Falcucci A, Ojima DS, Salvatore M, Conant RT, 2015. Greenhouse gas mitigation potential of the world’s grazing lands:

64

Modeling soil carbon and nitrogen fluxes of mitigation practices. Agric Ecosyst Environ 207, 91–100. doi:10.1016/j.agee.2015.03.029 Hoekstra AY, Chapagain AK, Aldaya MM, Mekonnen MM, 2011. The Water Footprint Assessment Manual Setting the Global Standard. Earthscan, London, Washington DC. pp. 228. Hoff H, 2011. Understanding the Nexus. Background Paper for the Bonn 2011 Conference: The Water, Energy and Food Security Nexus. Stockholm Environment Institute, Stockholm. Hua LM, Squires VR, 2015. Managing China’s pastoral lands: Current problems and future prospects. Land Use Policy 43, 129–137. doi:10.1016/j.landusepol.2014.11.004 IISD (International Institute for Sustainable Development), 2014. Coverage of Selected Side Events at the Lima Climate Change Conference - Building a Common Indicator on Land-Based Adaptation for Reporting on the Rio Conventions, 1-12 December. Available from:

http://www.iisd.ca/climate/cop20/enbots/11dec.html

IPCC (Intergovernmental Panel on Climate Change), 2000. Land Use, Land-Use Change and Forestry. Cambridge University Press, UK. pp 375. Available from: http://www.ipcc.ch/ipccreports/sres/land_use/index.php?idp=0, accessed 18 January 2015. IPCC (Intergovernmental Panel on Climate Change), 2006a. IPCC Guidelines for national Greenhouse Gas Inventories Vol. 4 Agriculture, Forestry and Other Land Use. IPCC/OECD/IEA/IGES, Hayama, Japan. Available from: http://www.ipcc-nggip.iges.or.jp/public/2006gl/index.html, accessed 1 July 2014. IPCC (Intergovernmental Panel on Climate Change), 2006b. IPCC Guidelines for national Greenhouse Gas Inventories Vol. 2 Energy. IPCC/OECD/IEA/IGES, Hayama, Japan. Available from: http://www.ipcc-nggip.iges.or.jp/public/2006gl/index.html, accessed 2 February 2015. IPCC (Intergovernmental Panel on Climate Change), 2007. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. Available from: http://www.ipcc.ch/publications_and_data/ar4/wg1/en/ch2s2-10-2.html, accessed 1 July 2014.

65

IPCC (Intergovernmental Panel on Climate Change), 2013. Climate Change 2013: The Physical Science Basis. Contribution of Working group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Available from: http://www.ipcc.ch/report/ar5/wg1/, accessed October 13 2014. IPCC (Intergovernmental Panel on Climate Change), 2014. Climate Change 2014: Mitigation of Climate Change. Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Available from: http://report.mitigation2014.org/drafts/final-draft-postplenary/ipcc_wg3_ar5_final-draft_postplenary_chapter11.pdf, accessed October 13 2014. Ji WH, Su JH, 2016. Range management: Tibetan wildlife hemmed in. Nature 530, 33. doi: 10.1038/530033c Kallis G, 2008. Droughts. Annu Rev Env Resour 33, 85–118. doi: 10.1146/annurev.environ.33.081307.123117 Karlberg L, 2014. Using a nexus approach to support development and environmental planning in Ethiopia. Discussion brief. Available from: http://www.sei-international.org/mediamanager/documents/Publications/Air-land-water-resources/SEI-DB-2014-Nexus-Blue-Nile.pdf, accessed 18 October 2014. Keys PW, van der Ent RJ, Gordon LJ, Hoff H, Nikoli R, Savenije HHG, 2012. Analyzing precipitationsheds to understand the vulnerability of rainfall dependent regions. Biogeosciences 9, 733-46. doi: 10.5194/bg-9-733-2012 Kleidon A, Fraedrich K, Heimann M, 2000. A Green Planet Versus a Desert World: Estimating the Maximum Effect of Vegetation on the Land Surface Climate. Clim Chang 44(4), 471-493. doi: 10.1023/A:1005559518889 Lal R, 2002. Soil carbon sequestration in china through agricultural intensification, and restoration of degraded and desertified ecosystems. Land Degrad Develop 13, 469–478. doi: 10.1002/ldr.531 Lal R, 2004a. Soil Carbon Sequestration Impacts on Global Climate Change and Food Security. Science 304(5677), 1623-1627. doi: 10.1126/science.1097396 Lal R, 2004b. Soil carbon sequestration to mitigate climate change. Geoderma 123(1–2), 1–22. doi: 10.1016/j.geoderma.2004.01.032 Liu C, Wu B, 2010. 'Grain for Green Program in China': Policy Making and Implementation? Briefing No. 60, China Policy Institute, University of Nottingham University. Avaialble from:

66

http://www.nottingham.ac.uk/cpi/documents/briefings/briefing-60-reforestation.pdf, accessed 14 October 2014. Moura-Costa P, Wilson C, 2000. An equivalence factor between CO2 avoidedemissions and sequestration – description andapplications in forestry. Mitig Adapt Strateg Glob Chang 5(1), 51-60. doi: 10.1023/A:1009697625521 Nachtergaele FO, Petri M, Biancalani R, van Lynden G, van Velthuizen H, Bloise M, 2011. LADA technical report n. 17 Global Land Degradation Information System (GLADIS) Version 1.0. An information database for land degradation assessment at global level. Available from: www.tinyurl.com/faogladis, accessed 16 December 2014. Ni J, 2002. Carbon storage in grasslands of China. J Arid Environ 50, 205–218 doi:10.1006/jare.2001.0902 Olivier JGJ, Janssens-Maenhout G, Muntean M, Peters JHAW, Trends in global CO2 emissions - 2013 report, PBL / JRC report 83593. EUR 26098 EN; ISBN 978-94-91506-51-2. Available from: http://edgar.jrc.ec.europa.eu/news_docs/pbl-2013-trends-in-global-co2-emissions-2013-report-1148.pdf, accessed 19 October 2014. Olsson A, Campana PE, Lind M, Yan JY, 2014. Potential for carbon sequestration and mitigation of climate by irrigation of grassland. Appl Energy 136, 1145–1154. doi: 10.1016/j.apenergy.2014.08.025 Olsson A, Campana PE, Lind M, Yan JY, 2015. PV Water Pumping for Carbon Sequestration in Dry Land Agriculture. Energy Convers Manag, Article in Press. doi:10.1016/j.enconman.2014.12.056 Ould-Amrouche S, Rekioua D, Hamidat A, 2010. Modelling photovoltaic water pumping systems and evaluation of their CO2 emissions mitigation potential. Appl Energy 87(11), 3451-3459. doi: 10.1016/j.apenergy.2010.05.021 Peng J, Lu L, Yang H, 2013. Review on life cycle assessment of energy pay back and greenhouse gas emission of solar photovoltaic systems. Renew Sust Energy Rev 19, 255–274. doi: 10.1016/j.rser.2012.11.035 Purohit P, Michaelowa A, 2008. CDM potential of SPV pumps in India. Renew Sust Energy Rev 12(1), 181–199. doi:10.1016/j.rser.2006.05.011 Qiu J, 2016. Trouble in Tibet. Nature 529, 142–145. doi:10.1038/529142a Roberts D, 2015. Creating a Desert in China – Beijing pressures herders to move to the cities [Internet]. Bloomberg Businessweek. Available from:

67

http://www.bloomberg.com/news/articles/2015-10-01/china-mongolian-desert-herders-under-pressure-to-move-to-cities, accessed 2 February 2016. Rockström, 2003. Water for food and nature in drought-prone tropics: vapour shift in rain-fed agriculture. Philos Trans R Soc Lond B Biol Sci 358(1440), 1997–2009. doi: 10.1098/rstb.2003.1400 Rockström J, Falkenmark M, Lannerstad M, Karlberg L, 2012. The planetary water drama: Dual task of feeding humanity and curbing climate change. Geophys Res Lett 39, L15401, 1-8. doi:10.1029/2012GL051688 Rockström J, Falkenmark M, Folke C, Lannerstad M, Barron J, Enfors E, Gordon L, Heinke J, Hoff H, Pahl-Wostl C, 2014. Water Resilience for Human Prosperity. Cambridge University Press, New York. pp. 311. Ryan JG, McAlpine CA, Ludwig JA, 2010. Integrated vegetation designs for enhancing water retention and recycling in agroecosystems. Landscape Ecol 25, 1277–1288. doi: 10.1007/s10980-010-9509-7 Savory Institute, 2014. Putting Grasslands to Work - Day 1 - Session 4. 42 min into the clip available from: https://www.youtube.com/watch?v=KbfYBeJDD68 Skinner RH, 2008. High biomass removal limits carbon sequestration potential of mature temperate pastures. J Environ Qual 37(4), 1319-1326. doi: 10.2134/jeq2007.0263

Smith WN, Grant BB, Campbell CA, McConkey BG, Desjardins RL, Kröbel R, Malhi SS, 2012. Crop residue removal effects on soil carbon: Measured and inter-model comparisons. Agric Ecosyst Environ 161, 27– 38. doi: 10.1016/j.agee.2012.07.024 SOU (Statens offentliga utredningar), 2013. SOU 2013:68 Synliggöra värdet av ekosystemtjänster – Åtgärder för välfärd genom biologisk mångfald och ekosystemtjänster [Swedish]. Stockmann U, Adams MA, Crawford JW, Field DJ, Henakaarchchi N, Jenkins M, Minasny B, McBratney AB, de Remy de Courcelles V, Singh K, Wheeler I, Abbott L, Angers DA, Baldock J, Bird M, Brookes PC, Chenu C, Jastrow JD, Lal R, Lehmann J, O’Donnell AG, Parton WJ, Whitehead D, Zimmermann M, 2013. The knowns, known unknowns and unknowns of sequestration of soil organic carbon. Agric Ecosyst Environ 164(0), 80-99 doi: 10.1016/j.agee.2012.10.001

68

Su YZ, 2007. Soil carbon and nitrogen sequestration following the conversion of cropland to alfalfa forage land in northwest China. Soil Tillage Res 92(1–2), 181–189. doi: 10.1016/j.still.2006.03.001 Tennigkeit T, Solymosi K, Seebauer M, Lager B, 2013. Carbon Intensification and Poverty Reduction in Kenya: Lessons from the Kenya Agricultural Carbon Project. Field Actions Sci Rep Special Issue 7. Available from: http://factsreports.revues.org/2600, accessed 19 January 2015. UN, 2014. Available from: http://www.un.org/en/events/desertification_decade/whynow.shtml, accessed 14 October 2014. UNCCD (United Nations Convention to Combat Desertification), 1994. Elaboration of an international convention to combat desertification in countries experiencing serious drought and/or desertification, particularly in Africa. A/AC.241/27. Available from: http://www.unccd.int/Lists/SiteDocumentLibrary/conventionText/conv-eng.pdf, accessed 19 October 2014. UNEP (United Nations Environment Programme), 2014. Risoe CDM/JI Pipeline Analysis and Database. Available at: http://cdmpipeline.org/cers.htm, accessed 8 January 2014. UNEP DTU (United Nations Environment Programme Danish Technical University), 2014. NAMA Pipeline Analysis and Database, July 1st 2014. Available from: http://www.namapipeline.org/, accessed 23 July 2014. UNFCCC (United Nations Framework Convention on Climate Change): 1992. United Nations Framework Convention on Climate Change. Climate Change Secretariat, Geneva, Switzerland. Available from: http://unfccc.int/resource/docs/convkp/conveng.pdf, accessed 19 October 2014. UNFCCC (United Nations Framework Convention on Climate Change), 1998. Kyoto Protocol to the United Nations Framework Convention on Climate Change. Adopted at COP3 in Kyoto, Japan, 1997 December 11. Climate Change Secretariat, Geneva, Switzerland. Available from: http://unfccc.int/resource/docs/convkp/kpeng.pdf, accessed 8 January 2014. UNFCCC (United Nations Framework Convention on Climate Change), 2006a. Indicative simplified baseline and monitoring methodologies for selected small-scale CDM project activity categories I.D./Version 9

69

UNFCCC (United Nations Framework Convention on Climate Change), 2006b. Decision 3/CMP.1. FCCC/KP/CMP/2005/8/Add.1. United Nations Office, Geneva, Switzerland. Available from: http://unfccc.int/resource/docs/2005/cmp1/eng/08a01.pdf, accessed 14 January 2015. UNFCCC (United Nations Framework Convention on Climate Change), 2006c. Decision 20/CMP.1. FCCC/KP/CMP/2005/8/Add.3. United Nations Office, Geneva, Switzerland. Available from: http://unfccc.int/resource/docs/2005/cmp1/eng/08a03.pdf, accessed 19 January 2015. UNFCCC (United Nations Framework Convention on Climate Change), 2008. Decision 1/CP.13. FCCC/CP/2007/6/Add.1. United Nations Office, Geneva, Switzerland. Available from: http://unfccc.int/resource/docs/2007/cop13/eng/06a01.pdf, accessed 14 January 2015. UNFCCC (United Nations Framework Convention on Climate Change), 2010. Decision 1/CP.16. FCCC/CP/2010/7/Add.1. United Nations Office, Geneva, Switzerland. Available from: http://unfccc.int/resource/docs/2010/cop16/eng/07a01.pdf, accessed 14 January 2015. UNFCCC (United Nations Framework Convention on Climate Change), 2011. EB 63 Report Annex 24. Available from: https://cdm.unfccc.int/filestorage/L/0/8/L08WMCZ5FO9ATRD7JP1X3BEHQK4NI2/eb63_repan24.pdf?t=TEF8bml5cjFnfDB_ank250q7uDzOK1mqQlPJ, accessed 30 January 2015. UNFCCC (United Nations Framework Convention on Climate Change), 2012a. Report of the Conference of the Parties serving as the meeting of the Parties to the Kyoto Protocol on its seventh session, held in Durban from 28 November to 11 December 2011. FCCC/KP/CMP/2011/10/Add.1 Available from: http://unfccc.int/resource/docs/2011/cmp7/eng/10a01.pdf, accessed 11 August 2014. UNFCCC (United Nations Framework Convention on Climate Change), 2012. AMS-I.D.: Grid connected renewable electricity generation, version 17.0. Available from: https://cdm.unfccc.int/methodologies/DB/RSCTZ8SKT4F7N1CFDXCSA7BDQ7FU1X, accessed 19 October 2014. UNFCCC (United Nations Framework Convention on Climate Change), 2014a. Project Cycle Search [Internet]. Project Design Documents of the 55 A/R CDM

70

projects. https://cdm.unfccc.int/Projects/projsearch.html, accessed 25 September 2014. UNFCCC (United Nations Framework Convention on Climate Change), 2014b. FCCC/TP/2014/2. Options for possible additional land use, land-use change and forestry activities and alternative approaches to addressing the risk of non-permanence under the clean development mechanism. Technical paper. Available at: http://unfccc.int/resource/docs/2014/tp/02.pdf, accessed 16 September 2014. UNFCCC (United Nations Framework Convention on Climate Change), 2014c. Nationally Appropriate Mitigation Actions (NAMAs) [Internet]. Available at: http://unfccc.int/focus/mitigation/items/7172txt.php, accessed 21 October 2014. UNFCCC (United Nations Framework Convention on Climate Change), 2015. UN’s CDM to be further simplified and streamlined [Internet]. Available from: https://cdm.unfccc.int/press/newsroom/latestnews/releases/2015/2805_index.html, accessed 3 June 2015. van der Ent RJ, Savenije HHG, Schaefli B, Steele-Dunne SC, 2010. Origin and fate of atmospheric moisture over continents. Water Resour Res 46(9), 1-12. VCS (Verified carbon standard), 2011. Approved VCS Methodology VM0017 Adoption of Sustainable Agricultural Land Management. Available from: http://www.v-c-s.org/methodologies/VM0017, accessed 19 October 2014. VCS (Verified carbon standard), 2012a. AFOLU Non-Permanence Risk Tool. Procedural Document v3.2. Available from: http://www.v-c-s.org/sites/v-c-s.org/files/AFOLU%20Non-Permanence%20Risk%20Tool,%20v3.2.pdf, accessed 10 June 2015. VCS (Verified carbon standard), 2012b. AFOLU non-permanence risk tool. Version 3.2. Available from: http://www.v-c-s.org/sites/v-c-s.org/files/AFOLU%20Non-Permanence%20Risk%20Tool,%20v3.2.pdf, accessed 19 January 2015. Wang XM, Chen FH, Hasi E, Li JC, 2008. Desertification in China: An assessment. Earth Sci Rev 88(3–4), 188–206. doi: 10.1016/j.earscirev.2008.02.001 Wang SP, Wilkes A, Zhang ZC, Chang XF, Dixon J, Dendooven L, 2011. Management and land use change effects on soil carbon in northern China’s grasslands: a synthesis. Agri Ecosyst Environ 142, 329–340. doi: 10.1016/j.agee.2011.06.002.

71

Wang F, Pan XB, Wang DF, Shen CY, Lu Q, 2013. Combating desertification in China: Past, present and future. Land Use Policy 31, 311–313. doi: 10.1016/j.landusepol.2012.07.010 Whitman T, Lehmann J, 2009. Biochar—One way forward for soil carbon in offset mechanisms in Africa? Environ Sci Policy 12 (7), 1024–1027. doi: 10.1016/j.envsci.2009.07.013 Wichelns D, 2015. Virtual water and water footprints do not provide helpful insight regarding international trade or water scarcity. Ecol Indic 52, 277–283. doi: 10.1016/j.ecolind.2014.12.013 Winkler H, Thorne S, 2002. Baselines for Suppressed Demand: CDM Projects Contribution to Poverty Alleviation. S Afr J Econ Manag Sci 5(2), 413-429. Available from: http://www.eri.uct.ac.za/Research/publications-pre2004/02Winkler-Thorne_Baselines_suppressed_demand.pdf, accessed 19 October 2014. WMO (World Meteorological Organization), 2014. [Internet] United Nations Framework Convention on Climate Change (UNFCCC). Available from: http://www.wmo.int/pages/themes/climate/international_unfccc.php, accessed 20 January 2015. World Bank, 2013. Guidance for the estimation of project removals due to changes in soil organic carbon with the RothC model. Available from: https://wbcarbonfinance.org/Router.cfm?Page=BioSALM, accessed 13 October 2014. Wu X, Li ZS, Fu BJ, Zhou WM, Liu HF, Liu GH, 2014. Restoration of ecosystem carbon and nitrogen storage and microbial biomass after grazing exclusion in semi-arid grasslands of Inner Mongolia. Ecol Eng 73, 395–403. doi: 10.1016/j.ecoleng.2014.09.077 Xiao JF, Sun G, Chen JQ, Chen H, Chen SP, Dong G, 2013. Carbon fluxes, evapotranspiration, and water use efficiency of terrestrial ecosystems in China. Agric For Meteorol 182–183, 76– 90. doi: 10.1016/j.agrformet.2013.08.007 Xie J, 2009. Addressing China’s Water Scarcity – Recommendations for Selected Water Resource Management Issues. The World Bank, Washington, DC, USA. Xiong DP, Shi PL, Sun YL, Wu JS, Zhang XZ, 2014. Effects of Grazing Exclusion on Plant Productivity and Soil Carbon, Nitrogen Storage in Alpine Meadows in Northern Tibet, China. Chin Geogra Sci 24(4), 488–498. doi: 10.1007/s11769-014-0697-y

72

Yeh ET, 2005. Green governmentality and pastoralism in Western China: ‘Converting pastures to grasslands’. Nomadic peoples 9(1-2), 9-30. Available from: http://tibet3rdpole.org/wp-content/uploads/nomadic-peoples-emily-yeh.pdf, accessed 19 October 2014. Yan J, Gao Z, Yu G, Campana PE, Liu J, Zhang C, Yang J, Olsson A, Li H, 2013. Demonstration and Scale-Up of Photovoltaic Solar Water Pumping for the Conservation of Grassland and Farmland in China. SIDA Project No.: AKT-2010-040. Zhang J, Liu JH, Campana PE, Zhang RQ, Yan JY, Gao XR, 2014a. Model of evapotranspiration and groundwater level based on photovoltaic water pumping system. Appl Energy 136, 1132–1137. doi:10.1016/j.apenergy.2014.05.045 Zhang J, Niu JM, Buyantuev A, Wu JG, 2014b. A multilevel analysis of effects of land use policy on land-cover change and local land use decisions. J Arid Environ 108, 19-28. doi: 10.1016/j.jaridenv.2014.04.006 Ziman, 2002. The continuing need for disinterested research. Sci Eng Ethics 8, 397-399. doi: 10.1007/s11948-002-0060-z Zurbrügg C, Volkart E, Hellweg S, 2012. Comparing LCA and CDM methods – two ways to calculate greenhouse gas emission reduction due to organic waste treatment. In: Proceedings of WasteSafe 2013, 3rd International Conference on Solid Waste Management in Developing Countries, Khulna, Bangladesh, 10-12 February 2013. Available from: http://www.eawag.ch/forschung/sandec/publikationen/swm/dl/lca_cdm.pdf, accessed October 13 2014.