Editing IPCC:AR6/SRCCL/Chapter-1 (section)

== CCB2 Implications of large-scale conversion from non-forest to forest land ==

<div id="section-1-3-1-targeted-decarbonisation-relying-on-large-land-area-need-block-1"></div>

Baldur Janz (Germany), Almut Arneth (Germany), Francesco Cherubini (Norway/Italy), Edouard Davin (Switzerland/France), Aziz Elbehri (Morocco), Kaoru Kitajima (Japan), Werner Kurz (Canada).

'''Efforts to increase forest area'''

While deforestation continues in many world regions, especially in the tropics, large expansion of mostly managed forest area has taken place in some countries. In the IPCC context, reforestation (conversion to forest of land that previously contained forests but has been converted to some other use) is distinguished from afforestation (conversion to forest of land that historically has not contained forests; see Glossary). Past expansion of managed forest area occurred in many world-regions for a variety of reasons, from meeting needs for wood fuel or timber (Vadell et al. 2016 <sup>[[#fn:r591|591]]</sup> ; Joshi et al. 2011 <sup>[[#fn:r592|592]]</sup> ; Zaloumis and Bond 2015 <sup>[[#fn:r593|593]]</sup> ; Payn et al. 2015 <sup>[[#fn:r594|594]]</sup> ; Shoyama 2008 <sup>[[#fn:r595|595]]</sup> ; Miyamoto et al. 2011 <sup>[[#fn:r596|596]]</sup> ) to restoration-driven efforts, with the aim of enhancing ecological function (Filoso et al. 2017 <sup>[[#fn:r597|597]]</sup> ; Salvati and Carlucci 2014 <sup>[[#fn:r598|598]]</sup> ; Ogle et al. 2018 <sup>[[#fn:r599|599]]</sup> ; Crouzeilles et al. 2016 <sup>[[#fn:r600|600]]</sup> ; FAO 2016 <sup>[[#fn:r601|601]]</sup> ) (Sections 3.7 and 4.9).

In many regions, net forest area increase includes deforestation (often of native forests) alongside increasing forest area (often managed forest, but also more natural forest restoration efforts) (Heilmayr et al. 2016 <sup>[[#fn:r602|602]]</sup> ; Scheidel and Work 2018 <sup>[[#fn:r603|603]]</sup> ; Hua et al. 2018 <sup>[[#fn:r604|604]]</sup> ; Crouzeilles et al. 2016 <sup>[[#fn:r605|605]]</sup> ; Chazdon et al. 2016b <sup>[[#fn:r606|606]]</sup> ). China and India have seen the largest net forest area increase, aiming to alleviate soil erosion, desertification and overgrazing (Ahrends et al. 2017 <sup>[[#fn:r607|607]]</sup> ; Cao et al. 2016 <sup>[[#fn:r608|608]]</sup> ; Deng et al. 2015 <sup>[[#fn:r609|609]]</sup> ; Chen et al. 2019 <sup>[[#fn:r610|610]]</sup> ) (Sections 3.7 and 4.9) but uncertainties in exact forest area changes remain large, mostly due to differences in methodology and forest classification (FAO 2015a <sup>[[#fn:r611|611]]</sup> ; Song et al. 2018 <sup>[[#fn:r612|612]]</sup> ; Hansen et al. 2013 <sup>[[#fn:r613|613]]</sup> ; MacDicken et al. 2015 <sup>[[#fn:r614|614]]</sup> ).

'''What are the implications for ecosystems?'''

''1. Implications for biogeochemical and biophysical processes''

There is robust evidence and medium agreement that whilst forest area expansion increases ecosystem carbon storage, the magnitude of the increased stock depends on the type and length of former land use, forest type planted, and climatic regions (Bárcena et al. 2014 <sup>[[#fn:r615|615]]</sup> ; Poeplau et al. 2011 <sup>[[#fn:r616|616]]</sup> ; Shi et al. 2013 <sup>[[#fn:r617|617]]</sup> ; Li et al. 2012 <sup>[[#fn:r618|618]]</sup> ) (Section 4.3). While reforestation of former croplands increases net ecosystem carbon storage (Bernal et al. 2018 <sup>[[#fn:r619|619]]</sup> ; Lamb 2018 <sup>[[#fn:r620|620]]</sup> ), afforestation on native grassland results in reduction of soil carbon stocks, which can reduce or negate the net carbon benefits which are dominated by increases in biomass, dead wood and litter carbon pools (Veldman et al. 2015, 2017 <sup>[[#fn:r621|621]]</sup> ).

Forest vs non-forest lands differ in land surface reflectiveness of shortwave radiation and evapotranspiration (Anderson et al. 2011 <sup>[[#fn:r622|622]]</sup> ; Perugini et al. 2017 <sup>[[#fn:r623|623]]</sup> ) (Section 2.4). Evapotranspiration from forests during the growing season regionally cools the land surface and enhances cloud cover that reduces shortwave radiation reaching the land, an impact that is especially pronounced in the tropics. However, dark evergreen conifer-dominated forests have low surface reflectance, and tend to cause warming of the near-surface atmosphere compared to non-forest land, especially when snow cover is present such as in boreal regions (Duveiller et al. 2018 <sup>[[#fn:r624|624]]</sup> ; Alkama and Cescatti 2016 <sup>[[#fn:r625|625]]</sup> ; Perugini et al. 2017 <sup>[[#fn:r626|626]]</sup> ) (medium evidence, high agreement).

''2. Implications for water balance''

Evapotranspiration by forests reduces surface runoff and erosion of soil and nutrients (Salvati et al. 2014 <sup>[[#fn:r627|627]]</sup> ). Planting of fast-growing species in semi-arid regions or replacing natural grasslands with forest plantations can divert soil water resources to evapotranspiration from groundwater recharge (Silveira et al. 2016 <sup>[[#fn:r628|628]]</sup> ; Zheng et al. 2016 <sup>[[#fn:r629|629]]</sup> ; Cao et al. 2016 <sup>[[#fn:r630|630]]</sup> ). Multiple cases are reported from China where afforestation programs, some with irrigation, without having been tailored to local precipitation conditions, resulted in water shortages and tree mortality (Cao et al. 2016; Yang et al. 2014 <sup>[[#fn:r631|631]]</sup> ; Li et al. 2014 <sup>[[#fn:r632|632]]</sup> ; Feng et al. 2016 <sup>[[#fn:r633|633]]</sup> ). Water shortages may create long-term water conflicts (Zheng et al. 2016 <sup>[[#fn:r634|634]]</sup> ). However, reforestation (in particular for restoration) is also associated with improved water filtration, groundwater recharge (Ellison et al. 2017 <sup>[[#fn:r635|635]]</sup> ) and can reduce risk of soil erosion, flooding, and associated disasters (Lee et al. 2018 <sup>[[#fn:r636|636]]</sup> ) (Section 4.9).

''3. Implications for biodiversity''

Impacts of forest area expansion on biodiversity depend mostly on the vegetation cover that is replaced: afforestation on natural non-tree-dominated ecosystems can have negative impacts on biodiversity (Abreu et al. 2017 <sup>[[#fn:r637|637]]</sup> ; Griffith et al. 2017 <sup>[[#fn:r638|638]]</sup> ; Veldman et al. 2015 <sup>[[#fn:r639|639]]</sup> ; Parr et al. 2014 <sup>[[#fn:r640|640]]</sup> ; Wilson et al. 2017 <sup>[[#fn:r641|641]]</sup> ; Hua et al. 2016 <sup>[[#fn:r642|642]]</sup> ; see also IPCC 1.5° report (2018)). Reforestation with monocultures of fast-growing, non-native trees has little benefit to biodiversity (Shimamoto et al. 2018 <sup>[[#fn:r643|643]]</sup> ; Hua et al. 2016). There are also concerns regarding some commonly used plantation species (e.g., Acacia and Pinus species) to become invasive (Padmanaba and Corlett 2014 <sup>[[#fn:r644|644]]</sup> ; Cunningham et al. 2015b <sup>[[#fn:r645|645]]</sup> ).

Reforestation with mixes of native species, especially in areas that retain fragments of native forest, can support ecosystem services and biodiversity recovery, with positive social and environmental co-benefits (Cunningham et al. 2015a <sup>[[#fn:r646|646]]</sup> ; Dendy et al. 2015 <sup>[[#fn:r647|647]]</sup> ; Chaudhary and Kastner 2016 <sup>[[#fn:r648|648]]</sup> ; Huang et al. 2018 <sup>[[#fn:r649|649]]</sup> ; Locatelli et al. 2015b <sup>[[#fn:r650|650]]</sup> ) (Section 4.5). Even though species diversity in re-growing forests is typically lower than in primary forests, planting native or mixed species can have positive effects on biodiversity (Brockerhoff et al. 2013 <sup>[[#fn:r651|651]]</sup> ; Pawson et al. 2013 <sup>[[#fn:r652|652]]</sup> ; Thompson et al. 2014 <sup>[[#fn:r653|653]]</sup> ). Reforestation has been shown to improve links among existing remnant forest patches, increasing species movement, and fostering gene flow between otherwise isolated populations (Gilbert-Norton et al. 2010 <sup>[[#fn:r654|654]]</sup> ; Barlow et al. 2007 <sup>[[#fn:r655|655]]</sup> ; Lindenmayer and Hobbs 2004 <sup>[[#fn:r656|656]]</sup> ).

''4. Implications for other ecosystem services and societies''

Forest area expansion could benefit recreation and health, preservation of cultural heritage and local values and knowledge, livelihood support (via reduced resource conflicts, restoration of local resources). These social benefits could be most successfully achieved if local communities’ concerns are considered (Le et al. 2012 <sup>[[#fn:r657|657]]</sup> ). However, these co-benefits have rarely been assessed due to a lack of suitable frameworks and evaluation tools (Baral et al. 2016 <sup>[[#fn:r658|658]]</sup> ).

Industrial forest management can be in conflict with the needs of forest-dependent people and community-based forest management over access to natural resources (Gerber 2011 <sup>[[#fn:r659|659]]</sup> ; Baral et al. 2016 <sup>[[#fn:r660|660]]</sup> ) and/or loss of customary rights over land use (Malkamäki et al. 2018 <sup>[[#fn:r661|661]]</sup> ; Cotula et al. 2014 <sup>[[#fn:r662|662]]</sup> ). A common result is out-migration from rural areas and diminishing local uses of ecosystems (Gerber 2011 <sup>[[#fn:r663|663]]</sup> ). Policies promoting large-scale tree plantations gain traction if these are reappraised in view of potential co-benefits with several ecosystem services and local societies (Bull et al. 2006 <sup>[[#fn:r664|664]]</sup> ; Le et al. 2012 <sup>[[#fn:r665|665]]</sup> ).

'''Scenarios of forest area expansion for land-based climate change mitigation'''

Conversion of non-forest to forest land has been discussed as a relatively cost-effective climate change mitigation option when compared to options in the energy and transport sectors (medium evidence, medium agreement) (de Coninck et al. 2018 <sup>[[#fn:r666|666]]</sup> ; Griscom et al. 2017 <sup>[[#fn:r667|667]]</sup> ; Fuss et al. 2018 <sup>[[#fn:r668|668]]</sup> ), and can have co-benefits with adaptation.

Sequestration of CO <sub>2</sub> from the atmosphere through forest area expansion has become a fundamental part of stringent climate change mitigation scenarios (Rogelj et al. 2018a <sup>[[#fn:r669|669]]</sup> ; Fuss et al. 2018 <sup>[[#fn:r670|670]]</sup> ) (e.g., Sections 2.5, 4.5 and 6.2). The estimated mitigation potential ranges from about 0.5 to 10 GtCO <sub>2</sub> yr–1 (robust evidence, medium agreement), and depends on assumptions regarding available land and forest carbon uptake potential (Houghton 2013 <sup>[[#fn:r671|671]]</sup> ; Houghton and Nassikas 2017 <sup>[[#fn:r672|672]]</sup> ; Griscom et al. 2017 <sup>[[#fn:r673|673]]</sup> ; Lenton 2014 <sup>[[#fn:r674|674]]</sup> ; Fuss et al. 2018 <sup>[[#fn:r675|675]]</sup> ; Smith 2016 <sup>[[#fn:r676|676]]</sup> ) (Section 2.5.1). In climate change mitigation scenarios, typically, no differentiation is made between reforestation and afforestation despite different overall environmental impacts between these two measures. Likewise, biodiversity conservation, impacts on water balances, other ecosystem services, or land-ownership – as constraints when simulating forest area expansion (Cross-Chapter Box 1 in Chapter 1) – tend not to be included as constraints when simulating forest area expansion.

Projected forest area increases, relative to today’s forest area, range from approximately 25% in 2050 and increase to nearly 50% by 2100 (Rogelj et al. 2018a <sup>[[#fn:r677|677]]</sup> ; Kreidenweis et al. 2016 <sup>[[#fn:r678|678]]</sup> ; Humpenoder et al. 2014 <sup>[[#fn:r679|679]]</sup> ). Potential adverse side-effects of such large-scale measures, especially for low-income countries, could be increasing food prices from the increased competition for land (Kreidenweis et al. 2016 <sup>[[#fn:r680|680]]</sup> ; Hasegawa et al. 2015 <sup>[[#fn:r681|681]]</sup> , 2018 <sup>[[#fn:r682|682]]</sup> ; Boysen et al. 2017 <sup>[[#fn:r683|683]]</sup> ) (Section 5.5). Forests also emit large amounts of biogenic volatile compounds that under some conditions contribute to the formation of atmospherically short-lived climate forcing compounds, which are also detrimental to health (Ashworth et al. 2013 <sup>[[#fn:r684|684]]</sup> ; Harrison et al. 2013 <sup>[[#fn:r685|685]]</sup> ). Recent analyses argued for an upper limit of about 5 million km2 of land globally available for climate change mitigation through reforestation, mostly in the tropics (Houghton 2013 <sup>[[#fn:r686|686]]</sup> ) – with potential regional co-benefits.

Since forest growth competes for land with bioenergy crops (Harper et al. 2018 <sup>[[#fn:r687|687]]</sup> ) (Cross-Chapter Box 7 in Chapter 6), global area estimates need to be assessed in light of alternative mitigation measures at a given location. In all forest-based mitigation efforts, the sequestration potential will eventually saturate unless the area keeps expanding, or harvested wood is either used for long-term storage products or for carbon capture and storage (Fuss et al. 2018 <sup>[[#fn:r688|688]]</sup> ; Houghton et al. 2015 <sup>[[#fn:r689|689]]</sup> ) (Section 2.5.1). Considerable uncertainty in forest carbon uptake estimates is further introduced by potential forest losses from fire or pest outbreaks (Allen et al. 2010 <sup>[[#fn:r690|690]]</sup> ; Anderegg et al. 2015 <sup>[[#fn:r691|691]]</sup> ) (Cross-Chapter Box 3 in Chapter 2). And like all land-based mitigation measures, benefits may be diminshed by land-use displacement, and through trade of land-based products, especially in poor countries that experience forest loss (e.g., Africa) (Bhojvaid et al. 2016 <sup>[[#fn:r692|692]]</sup> ; Jadin et al. 2016 <sup>[[#fn:r693|693]]</sup> ).

'''Conclusion'''

Reforestation is a mitigation measure with potential co-benefits for conservation and adaptation, including biodiversity habitat, air and water filtration, flood control, enhanced soil fertility and reversal of land degradation. Potential adverse side-effects of forest area expansion depend largely on the state of the land it displaces as well as tree species selections. Active governance and planning contribute to maximising co-benefits while minimising adverse side-effects (Laestadius et al. 2011 <sup>[[#fn:r694|694]]</sup> ; Dinerstein et al. 2015 <sup>[[#fn:r695|695]]</sup> ; Veldman et al. 2017 <sup>[[#fn:r696|696]]</sup> ) (Section 4.8 and Chapter 7). At large spatial scales, forest expansion is expected to lead to increased competition for land, with potentially undesirable impacts on food prices, biodiversity, non-forest ecosystems and water availability (Bryan and Crossman 2013 <sup>[[#fn:r697|697]]</sup> ; Boysen et al. 2017 <sup>[[#fn:r698|698]]</sup> ; Kreidenweis et al. 2016 <sup>[[#fn:r699|699]]</sup> ; Egginton et al. 2014 <sup>[[#fn:r700|700]]</sup> ; Cao et al. 2016 <sup>[[#fn:r701|701]]</sup> ; Locatelli et al. 2015a <sup>[[#fn:r702|702]]</sup> ; Smith et al. 2013 <sup>[[#fn:r703|703]]</sup> ).

<span id="land-management"></span>
=== 1.3.2 Land management ===

<div id="section-1-3-2-1-agricultural-forest-and-soil-management"></div>

<span id="agricultural-forest-and-soil-management"></span>
==== 1.3.2.1 Agricultural, forest and soil management ====

<div id="section-1-3-2-1-agricultural-forest-and-soil-management-block-1"></div>

Sustainable land management (SLM) describes “the stewardship and use of land resources, including soils, water, animals and plants, to meet changing human needs while simultaneously assuring the long-term productive potential of these resources and the maintenance of their environmental functions” (Alemu 2016 <sup>[[#fn:r704|704]]</sup> ; Altieri and Nicholls 2017 <sup>[[#fn:r705|705]]</sup> ) (e.g., Section 4.1.5), and includes ecological, technological and governance aspects.

The choice of SLM strategy is a function of regional context and land-use types, with ''high agreement'' on (a combination of) choices such as agroecology (including agroforestry), conservation agriculture and forestry practices, crop and forest species diversity, appropriate crop and forest rotations, organic farming, integrated pest management, the preservation and protection of pollination services, rainwater harvesting, range and pasture management, and precision agriculture systems (Stockmann et al. 2013 <sup>[[#fn:r706|706]]</sup> ; Ebert, 2014 <sup>[[#fn:r707|707]]</sup> ; Schulte et al. 2014 <sup>[[#fn:r708|708]]</sup> ; Zhang et al. 2015 <sup>[[#fn:r709|709]]</sup> ; Sunil and Pandravada 2015 <sup>[[#fn:r710|710]]</sup> ; Poeplau and Don 2015 <sup>[[#fn:r711|711]]</sup> ; Agus et al. 2015 <sup>[[#fn:r712|712]]</sup> ; Keenan 2015 <sup>[[#fn:r713|713]]</sup> ; MacDicken et al. 2015 <sup>[[#fn:r714|714]]</sup> ; Abberton et al. 2016 <sup>[[#fn:r715|715]]</sup> ). Conservation agriculture and forestry uses management practices with minimal soil disturbance such as no tillage or minimum tillage, permanent soil cover with mulch, combined with rotations to ensure a permanent soil surface, or rapid regeneration of forest following harvest (Hobbs et al. 2008 <sup>[[#fn:r716|716]]</sup> ; Friedrich et al. 2012 <sup>[[#fn:r717|717]]</sup> ). Vegetation and soils in forests and woodland ecosystems play a crucial role in regulating critical ecosystem processes, therefore reduced deforestation together with sustainable forest management are integral to SLM (FAO 2015b <sup>[[#fn:r718|718]]</sup> ) (Section 4.8). In some circumstances, increased demand for forest products can also lead to increased management of carbon storage in forests (Favero and Mendelsohn 2014 <sup>[[#fn:r719|719]]</sup> ). Precision agriculture is characterised by a “management system that is information and technology based, is site specific and uses one or more of the following sources of data: soils, crops, nutrients, pests, moisture, or yield, for optimum profitability, sustainability, and protection of the environment” (USDA 2007 <sup>[[#fn:r720|720]]</sup> ) (Cross-Chapter Box 6 in Chapter 5). The management of protected areas that reduce deforestation also plays an important role in climate change mitigation and adaptation while delivering numerous ecosystem services and sustainable development benefits (Bebber and Butt 2017 <sup>[[#fn:r721|721]]</sup> ). Similarly, when managed in an integrated and sustainable way, peatlands are also known to provide numerous ecosystem services, as well as socio-economic and mitigation and adaptation benefits (Ziadat et al. 2018 <sup>[[#fn:r722|722]]</sup> ).

Biochar is an organic compound used as soil amendment and is believed to be potentially an important global resource for mitigation. Enhancing the carbon content of soil and/or use of biochar (Chapter 4) have become increasingly important as a climate change mitigation option with possibly large co-benefits for other ecosystem services. Enhancing soil carbon storage and the addition of biochar can be practiced with limited competition for land, provided no productivity/ yield loss and abundant unused biomass, but evidence is limited and impacts of large scale application of biochar on the full GHG balance of soils, or human health are yet to be explored (Gurwick et al. 2013 <sup>[[#fn:r723|723]]</sup> ; Lorenz and Lal 2014 <sup>[[#fn:r724|724]]</sup> ; Smith 2016 <sup>[[#fn:r725|725]]</sup> ).

<span id="value-chain-management"></span>
=== 1.3.3 Value chain management ===

<div id="section-1-3-3-1-supply-management"></div>

<span id="supply-management"></span>
==== 1.3.3.1 Supply management ====

<div id="section-1-3-3-1-supply-management-block-1"></div>

Food losses from harvest to retailer. Approximately one-third of losses and waste in the food system occurs between crop production and food consumption, increasing substantially if losses in livestock production and overeating are included (Gustavsson et al. 2011 <sup>[[#fn:r726|726]]</sup> ; Alexander et al. 2017 <sup>[[#fn:r727|727]]</sup> ). This includes on-farm losses, farm to retailer losses, as well retailer and consumer losses (Section 1.3.3.2).

Post-harvest food loss – on farm and from farm to retailer – is a widespread problem, especially in developing countries (Xue et al. 2017 <sup>[[#fn:r728|728]]</sup> ), but are challenging to quantify. For instance, averaged for eastern and southern Africa an estimated 10–17% of annual grain production is lost (Zorya et al. 2011 <sup>[[#fn:r729|729]]</sup> ). Across 84 countries and different time periods, annual median losses in the supply chain before retailing were estimated at about 28 kg per capita for cereals or about 12 kg per capita for eggs and dairy products (Xue et al. 2017 <sup>[[#fn:r730|730]]</sup> ). For the year 2013, losses prior to the reaching retailers were estimated at 20% (dry weight) of the production amount (22% wet weight) (Gustavsson et al. 2011 <sup>[[#fn:r731|731]]</sup> ; Alexander et al. 2017 <sup>[[#fn:r732|732]]</sup> ). While losses of food cannot be realistically reduced to zero, advancing harvesting technologies (Bradford et al. 2018 <sup>[[#fn:r733|733]]</sup> ; Affognon et al. 2015 <sup>[[#fn:r734|734]]</sup> ), storage capacity (Chegere 2018 <sup>[[#fn:r735|735]]</sup> ) and efficient transportation could all contribute to reducing these losses with co-benefits for food availability, the land area needed for food production and related GHG emissions.

'''Stability of food supply, transport and distribution.''' Increased climate variability enhances fluctuations in world food supply and price variability (Warren 2014 <sup>[[#fn:r736|736]]</sup> ; Challinor et al. 2015 <sup>[[#fn:r737|737]]</sup> ; Elbehri et al. 2017 <sup>[[#fn:r738|738]]</sup> ). ‘Food price shocks’ need to be understood regarding their transmission across sectors and borders and impacts on poor and food insecure populations, including urban poor subject to food deserts and inadequate food accessibility (Widener et al. 2017 <sup>[[#fn:r739|739]]</sup> ; Lehmann et al. 2013 <sup>[[#fn:r740|740]]</sup> ; Le 2016 <sup>[[#fn:r741|741]]</sup> ; FAO 2015b <sup>[[#fn:r742|742]]</sup> ). Trade can play an important stabilising role in food supply, especially for regions with agro-ecological limits to production, including water scarce regions, as well as regions that experience short-term production variability due to climate, conflicts or other economic shocks (Gilmont 2015 <sup>[[#fn:r743|743]]</sup> ; Marchand et al. 2016 <sup>[[#fn:r744|744]]</sup> ). Food trade can either increase or reduce the overall environmental impacts of agriculture (Kastner et al. 2014 <sup>[[#fn:r745|745]]</sup> ). Embedded in trade are virtual transfers of water, land area, productivity, ecosystem services, biodiversity, or nutrients (Marques et al. 2019 <sup>[[#fn:r746|746]]</sup> ; Wiedmann and Lenzen 2018 <sup>[[#fn:r747|747]]</sup> ; Chaudhary and Kastner 2016 <sup>[[#fn:r748|748]]</sup> ) with either positive or negative implications (Chen et al. 2018 <sup>[[#fn:r749|749]]</sup> ; Yu et al. 2013 <sup>[[#fn:r750|750]]</sup> ). Detrimental consequences in countries in which trade dependency may accentuate the risk of food shortages from foreign production shocks could be reduced by increasing domestic reserves or importing food from a diversity of suppliers (Gilmont 2015 <sup>[[#fn:r751|751]]</sup> ; Marchand et al. 2016 <sup>[[#fn:r752|752]]</sup> ).

Climate mitigation policies could create new trade opportunities (e.g., biomass) (Favero and Massetti 2014 <sup>[[#fn:r753|753]]</sup> ) or alter existing trade patterns. The transportation GHG footprints of supply chains may be causing a differentiation between short and long supply chains (Schmidt et al. 2017 <sup>[[#fn:r754|754]]</sup> ) that may be influenced by both economics and policy measures (Section 5.4). In the absence of sustainable practices and when the ecological footprint is not valued through the market system, trade can also exacerbate resource exploitation and environmental leakages, thus weakening trade mitigation contributions (Dalin and Rodríguez-Iturbe 2016 <sup>[[#fn:r755|755]]</sup> ; Mosnier et al. 2014 <sup>[[#fn:r756|756]]</sup> ; Elbehri et al. 2017 <sup>[[#fn:r757|757]]</sup> ). Ensuring stable food supply while pursuing climate mitigation and adaptation will benefit from evolving trade rules and policies that allow internalisation of the cost of carbon (and costs of other vital resources such as water, nutrients). Likewise, future climate change mitigation policies would gain from measures designed to internalise the environmental costs of resources and the benefits of ecosystem services (Elbehri et al. 2017 <sup>[[#fn:r758|758]]</sup> ; Brown et al. 2007 <sup>[[#fn:r759|759]]</sup> ).

<div id="section-1-3-3-2-demand-management"></div>

<span id="demand-management"></span>
==== 1.3.3.2 Demand management ====

<div id="section-1-3-3-2-demand-management-block-1"></div>

'''Dietary change.''' Demand-side solutions to climate mitigation are an essential complement to supply-side, technology and productivity driven solutions ( ''high confidence'' ) (Creutzig et al. 2016 <sup>[[#fn:r760|760]]</sup> ; Bajželj et al. 2014 <sup>[[#fn:r761|761]]</sup> ; Erb et al. 2016b <sup>[[#fn:r762|762]]</sup> ; Creutzig et al. 2018 <sup>[[#fn:r763|763]]</sup> ) (Sections 5.5.1 and 5.5.2). The environmental impacts of the animal-rich ‘western diets’ are being examined critically in the scientific literature (Hallström et al. 2015 <sup>[[#fn:r764|764]]</sup> ; Alexander et al. 2016b <sup>[[#fn:r765|765]]</sup> ; Alexander et al. 2015 <sup>[[#fn:r766|766]]</sup> ; Tilman and Clark 2014 <sup>[[#fn:r767|767]]</sup> ; Aleksandrowicz et al. 2016 <sup>[[#fn:r768|768]]</sup> ; Poore and Nemecek 2018 <sup>[[#fn:r769|769]]</sup> ) (Section 5.4.6). For example, if the average diet of each country were consumed globally, the agricultural land area needed to supply these diets would vary 14-fold, due to country differences in ruminant protein and calorific intake (–55% to +178% compared to existing cropland areas). Given the important role enteric fermentation plays in methane (CH4) emissions, a number of studies have examined the implications of lower animal-protein diets (Swain et al. 2018 <sup>[[#fn:r770|770]]</sup> ; Röös et al. 2017 <sup>[[#fn:r771|771]]</sup> ; Rao et al. 2018 <sup>[[#fn:r772|772]]</sup> ). Reduction of animal protein intake has been estimated to reduce global green water (from precipitation) use by 11% and blue water (from rivers, lakes, groundwater) use by 6% (Jalava et al. 2014 <sup>[[#fn:r773|773]]</sup> ). By avoiding meat from producers with above-median GHG emissions and halving animal-product intake, consumption change could free-up 21 million km <sup>2</sup> of agricultural land and reduce GHG emissions by nearly 5 GtCO <sub>2</sub> -eq yr <sup>–1</sup> or up to 10.4 GtCO <sub>2</sub> -eq yr <sup>–1</sup> when vegetation carbon uptake is considered on the previously agricultural land (Poore and Nemecek 2018 <sup>[[#fn:r774|774]]</sup> , 2019).

Diets can be location and community specific, are rooted in culture and traditions while responding to changing lifestyles driven for instance by urbanisation and changing income. Changing dietary and consumption habits would require a combination of non-price (government procurement, regulations, education and awareness raising) and price incentives (Juhl and Jensen 2014 <sup>[[#fn:r775|775]]</sup> ) to induce consumer behavioural change with potential synergies between climate, health and equity (addressing growing global nutrition imbalances that emerge as undernutrition, malnutrition, and obesity) (FAO 2018b <sup>[[#fn:r776|776]]</sup> ).

'''Reduced waste and losses in the food demand system.''' Global averaged per capita food waste and loss (FWL) have increased by 44% between 1961 and 2011 (Porter et al. 2016 <sup>[[#fn:r777|777]]</sup> ) and are now around 25–30% of global food produced (Kummu et al. 2012 <sup>[[#fn:r778|778]]</sup> ; Alexander et al. 2017 <sup>[[#fn:r779|779]]</sup> ). Food waste occurs at all stages of the food supply chain from the household to the marketplace (Parfitt et al. 2010 <sup>[[#fn:r780|780]]</sup> ) and is found to be larger at household than at supply chain levels. A meta-analysis of 55 studies showed that the highest share of food waste was at the consumer stage (43.9% of total) with waste increasing with per capita GDP for high-income countries until a plateaux at about 100 kg cap <sup>–1</sup> yr <sup>–1</sup> (around 16% of food consumption) above about 70,000 USD cap <sup>–1</sup> (van der Werf and Gilliland 2017 <sup>[[#fn:r781|781]]</sup> ; Xue et al. 2017 <sup>[[#fn:r782|782]]</sup> ). Food loss from supply chains tends to be more prevalent in less developed countries where inadequate technologies, limited infrastructure, and imperfect markets combine to raise the share of the food production lost before use.

There are several causes behind food waste including economics (cheap food), food policies (subsidies) as well as individual behaviour (Schanes et al. 2018 <sup>[[#fn:r783|783]]</sup> ). Household level food waste arises from overeating or overbuying (Thyberg and Tonjes 2016 <sup>[[#fn:r784|784]]</sup> ). Globally, overconsumption was found to waste 9–10% of food bought (Alexander et al. 2017 <sup>[[#fn:r785|785]]</sup> ).

Solutions to FWL thus need to address technical and economic aspects. Such solutions would benefit from more accurate data on the loss-source, loss-magnitude and causes along the food supply chain. In the long run, internalising the cost of food waste into the product price would more likely induce a shift in consumer behaviour towards less waste and more nutritious, or alternative, food intake (FAO 2018b <sup>[[#fn:r786|786]]</sup> ). Reducing FWL would bring a range of benefits for health, reducing pressures on land, water and nutrients, lowering emissions and safeguarding food security. Reducing food waste by 50% would generate net emissions reductions in the range of 20 to 30% of total food-sourced GHGs (Bajželj et al. 2014 <sup>[[#fn:r787|787]]</sup> ). SDG 12 (“Ensure sustainable consumption and production patterns”) calls for per capita global food waste to be reduced by one-half at the retail and consumer level, and reducing food losses along production and supply chains by 2030.

<span id="risk-management"></span>
=== 1.3.4 Risk management ===

<div id="section-1-3-4-risk-management-block-1"></div>

Risk management refers to plans, actions, strategies or policies to reduce the likelihood and/or magnitude of adverse potential consequences, based on assessed or perceived risks. Insurance and early warning systems are examples of risk management, but risk can also be reduced (or resilience enhanced) through a broad set of options ranging from seed sovereignty, livelihood diversification, to reducing land loss through urban sprawl. Early warning systems support farmer decision-making on management strategies (Section 1.2) and are a good example of an adaptation measure with mitigation co-benefits such as reducing carbon losses (Section 1.3.6). Primarily designed to avoid yield losses, early warning systems also support fire management strategies in forest ecosystems, which prevents financial as well as carbon losses (de Groot et al. 2015 <sup>[[#fn:r788|788]]</sup> ). Given that over recent decades on average around 10% of cereal production was lost through extreme weather events (Lesk et al. 2016 <sup>[[#fn:r790|790]]</sup> ), where available and affordable, insurance can buffer farmers and foresters against the financial losses incurred through such weather and other (fire, pests) extremes (Falco et al. 2014 <sup>[[#fn:r791|791]]</sup> ) (Sections 7.2 and 7.4). Decisions to take up insurance are influenced by a range of factors such as the removal of subsidies or targeted education (Falco et al. 2014). Enhancing access and affordability of insurance in low-income countries is a specific objective of the UNFCCC (Linnerooth-Bayer and Mechler 2006 <sup>[[#fn:r792|792]]</sup> ). A global mitigation co-benefit of insurance schemes may also include incentives for future risk reduction (Surminski and Oramas-Dorta 2014 <sup>[[#fn:r793|793]]</sup> ).

<span id="economics-of-land-based-mitigation-pathways-costs-versus-benefits-of-early-action-under-uncertainty"></span>
=== 1.3.5 Economics of land-based mitigation pathways: Costs versus benefits of early action under uncertainty ===

<div id="section-1-3-5-economics-of-land-based-mitigation-pathways-costs-versus-benefits-of-early-action-under-uncertainty-block-1"></div>

The overarching societal costs associated with GHG emissions and the potential implications of mitigation activities can be measured by various metrics (cost-benefit analysis, cost effectiveness analysis) at different scales (project, technology, sector or the economy) (IPCC 2018 <sup>[[#fn:r794|794]]</sup> ) (Section 1.4). The social cost of carbon (SCC) measures the total net damages of an extra metric tonne of CO <sub>2</sub>  emissions due to the associated climate change (Nordhaus 2014 <sup>[[#fn:r795|795]]</sup> ; Pizer et al. 2014 <sup>[[#fn:r796|796]]</sup> ). Both negative and positive impacts are monetised and discounted to arrive at the net value of consumption loss. As the SCC depends on discount rate assumptions and value judgements (e.g., relative weight given to current vs future generations), it is not a straightforward policy tool to compare alternative options. At the sectoral level, marginal abatement cost curves (MACCs) are widely used for the assessment  of costs related to GHG emissions reduction. MACCs measure the cost of reducing one more GHG unit and are either expert-based or model-derived and offer a range of approaches and assumptions on discount rates or available abatement technologies (Kesicki 2013 <sup>[[#fn:r797|797]]</sup> ). In land-based sectors, Gillingham and Stock (2018) <sup>[[#fn:r798|798]]</sup> reported short-term static abatement costs for afforestation of between 1 and 10 USD2017 per tCO <sub>2</sub> , soil management at 57 and livestock management at 71 USD2017 per tCO <sub>2</sub> . MACCs are more reliable when used to rank alternative options compared to a baseline (or business as usual) rather than offering absolute numerical measures (Huang et al. 2016 <sup>[[#fn:r799|799]]</sup> ). The economics of land-based mitigation options encompass also the “costs of inaction” that arise either from the economic damages due to continued accumulation of GHGs in the atmosphere and from the diminution in value of ecosystem services or the cost of their restoration where feasible (Rodriguez-Labajos 2013 <sup>[[#fn:r800|800]]</sup> ; Ricke et al. 2018 <sup>[[#fn:r801|801]]</sup> ). Overall, it remains challenging to estimate the costs of alternative mitigation options owing to the context – and scale-specific interplay between multiple drivers (technological, economic, and socio-cultural) and enabling policies and institutions (IPCC 2018 <sup>[[#fn:r802|802]]</sup> ) (Section 1.4).

The costs associated with mitigation (both project-linked such as capital costs or land rental rates, or sometimes social costs) generally increase with stringent mitigation targets and over time. Sources of uncertainty include the future availability, cost and performance of technologies (Rosen and Guenther 2015 <sup>[[#fn:r803|803]]</sup> ; Chen et al. 2016 <sup>[[#fn:r804|804]]</sup> ) or lags in decision-making, which have been demonstrated by the uptake of land use and land utilisation policies (Alexander et al. 2013 <sup>[[#fn:r805|805]]</sup> ; Hull et al. 2015 <sup>[[#fn:r806|806]]</sup> ; Brown et al. 2018b <sup>[[#fn:r807|807]]</sup> ). There is growing evidence of significant mitigation gains through conservation, restoration and improved land management practices (Griscom et al. 2017 <sup>[[#fn:r808|808]]</sup> ; Kindermann et al. 2008 <sup>[[#fn:r809|809]]</sup> ; Golub et al. 2013 <sup>[[#fn:r810|810]]</sup> ; Favero et al. 2017 <sup>[[#fn:r811|811]]</sup> ) (Chapters 4 and 6), but the mitigation cost efficiency can vary according to region and specific ecosystem (Albanito et al. 2016 <sup>[[#fn:r812|812]]</sup> ). Recent model developments that treat process-based, human–environment interactions have recognised feedbacks that reinforce or dampen the original stimulus for land-use change (Robinson et al. 2017 <sup>[[#fn:r813|813]]</sup> ; Walters and Scholes 2017 <sup>[[#fn:r814|814]]</sup> ). For instance, land mitigation interventions that rely on large-scale, land-use change (e.g., afforestation) would need to account for the rebound effect (which dampens initial impacts due to feedbacks) in which raising land prices also raises the cost of land-based mitigation (Vivanco et al. 2016 <sup>[[#fn:r815|815]]</sup> ). Although there are few direct estimates, indirect assessments strongly point to much higher costs if action is delayed or limited in scope ( ''medium confidence'' ). Quicker response options are also needed to avoid loss of high-carbon ecosystems and other vital ecosystem services that provide multiple services that are difficult to replace (peatlands, wetlands, mangroves, forests) (Yirdaw et al. 2017 <sup>[[#fn:r816|816]]</sup> ; Pedrozo-Acuña et al. 2015 <sup>[[#fn:r817|817]]</sup> ). Delayed action would raise relative costs in the future or could make response options less feasible ( ''medium confidence'' ) (Goldstein et al. 2019 <sup>[[#fn:r818|818]]</sup> ; Butler et al. 2014 <sup>[[#fn:r819|819]]</sup> ).

<span id="adaptation-measures-and-scope-for-co-benefits-with-mitigation"></span>
=== 1.3.6 Adaptation measures and scope for co-benefits with mitigation ===

<div id="section-1-3-6-adaptation-measures-and-scope-for-co-benefits-with-mitigation-block-1"></div>

Adaptation and mitigation have generally been treated as two separate discourses, both in policy and practice, with mitigation addressing cause and adaptation dealing with the consequences of climate change (Hennessey et al. 2017 <sup>[[#fn:r820|820]]</sup> ). While adaptation (e.g., reducing flood risks) and mitigation (e.g., reducing non-CO <sub>2</sub> emissions from agriculture) may have different objectives and operate at different scales, they can also generate joint outcomes (Locatelli et al. 2015b <sup>[[#fn:r821|821]]</sup> ) with adaptation generating mitigation co-benefits. Seeking to integrate strategies for achieving adaptation and mitigation goals is attractive in order to reduce competition for limited resources and trade-offs (Lobell et al. 2013 <sup>[[#fn:r822|822]]</sup> ; Berry et al. 2015 <sup>[[#fn:r823|823]]</sup> ; Kongsager and Corbera 2015 <sup>[[#fn:r824|824]]</sup> ). Moreover, determinants that can foster adaptation and mitigation practices are similar. These tend to include available technology and resources, and credible information for policymakers to act on (Yohe 2001 <sup>[[#fn:r825|825]]</sup> ).

Four sets of mitigation–adaptation interrelationships can be distinguished: (i) mitigation actions that can result in adaptation benefits; (ii) adaptation actions that have mitigation benefits; (iii) processes that have implications for both adaptation and mitigation; and (iv) strategies and policy processes that seek to promote an integrated set of responses for both adaptation and mitigation (Klein et al. 2007). A high level of adaptive capacity is a key ingredient to developing successful mitigation policy. Implementing mitigation action can result in increasing resilience especially if it is able to reduce risks. Yet, mitigation and adaptation objectives, scale of implementation, sector and even metrics to identify impacts tend to differ (Ayers and Huq 2009 <sup>[[#fn:r826|826]]</sup> ), and institutional setting, often does not enable an environment where synergies are sought (Kongsager et al. 2016 <sup>[[#fn:r827|827]]</sup> ). Trade-offs between adaptation and mitigation exist as well and need to be understood (and avoided) to establish win-win situations (Porter et al. 2014 <sup>[[#fn:r828|828]]</sup> ; Kongsager et al. 2016 <sup>[[#fn:r829|829]]</sup> ).

Forestry and agriculture offer a wide range of lessons for the integration of adaptation and mitigation actions given the vulnerability of forest ecosystems or cropland to climate variability and change (Keenan 2015 <sup>[[#fn:r830|830]]</sup> ; Gaba et al. 2015 <sup>[[#fn:r831|831]]</sup> ) (Sections 5.6 and 4.8). Increasing adaptive capacity in forested areas has the potential to prevent deforestation and forest degradation (Locatelli et al. 2011 <sup>[[#fn:r832|832]]</sup> ). Reforestation projects, if well managed, can increase community economic opportunities that encourage conservation (Nelson and de Jong 2003 <sup>[[#fn:r833|833]]</sup> ), build capacity through training of farmers and installation of multifunctional plantations with income generation (Reyer et al. 2009 <sup>[[#fn:r834|834]]</sup> ), strengthen local institutions (Locatelli et al. 2015a <sup>[[#fn:r835|835]]</sup> ) and increase cash-flow to local forest stakeholders from foreign donors (West 2016 <sup>[[#fn:r836|836]]</sup> ). A forest plantation that sequesters carbon for mitigation can also reduce water availability to downstream populations and heighten their vulnerability to drought. Inversely, not recognising mitigation in adaptation projects may yield adaptation measures that increase greenhouse gas emissions, a prime example of ‘maladaptation’. Analogously, ‘mal-mitigation’ would result in reducing GHG emissions, but increasing vulnerability (Barnett and O’Neill 2010 <sup>[[#fn:r837|837]]</sup> ; Porter et al. 2014 <sup>[[#fn:r838|838]]</sup> ). For instance, the cost of pursuing large-scale adaptation and mitigation projects has been associated with higher failure risks, onerous transactions costs and the complexity of managing big projects (Swart and Raes 2007 <sup>[[#fn:r839|839]]</sup> ).

Adaptation encompasses both biophysical and socio-economic vulnerability and underlying causes (informational, capacity, financial, institutional, and technological; Huq et al. 2014 <sup>[[#fn:r840|840]]</sup> ) and it is increasingly linked to resilience and to broader development goals (Huq et al. 2014 <sup>[[#fn:r841|841]]</sup> ). Adaptation measures can increase performance of mitigation projects under climate change and legitimise mitigation measures through the more immediately felt effects of adaptation (Locatelli et al. 2011 <sup>[[#fn:r842|842]]</sup> ; Campbell et al. 2014 <sup>[[#fn:r843|843]]</sup> ; Locatelli et al. 2015b <sup>[[#fn:r844|844]]</sup> ). Effective climate policy integration in the land sector is expected to gain from (i) internal policy coherence between adaptation and mitigation objectives, (ii) external climate coherence between climate change and development objectives, (iii) policy integration that favours vertical governance structures to foster effective mainstreaming of climate change into sectoral policies, and (iv) horizontal policy integration through overarching governance structures to enable cross-sectoral coordination (Sections 1.4 and 7.4).

<span id="enabling-the-response"></span>