Risk management and decision making in relation to sustainable development

There is high confidence that climate change has resulted in decreases in yield (of wheat, rice, maize, soy) and reduced food availability in low-latitude regions (IPCC, 2018⁴⁶) (Section 5.2.2). Countries in low- latitude regions are particularly vulnerable because the livelihoods of high proportions of the population are dependent on agricultural production. Even moderate temperature increases (1°C to 2°C) have negative yield impacts for major cereals, because the climate of many tropical agricultural regions is already quite close to the high-temperature thresholds for suitable production of these cereals (Rosenzweig et al. 2014⁴⁷). Thus, by 1.5°C global mean temperature GMT, or between approximately 1.6°C and approximately 2.6°C of local warming, risks to yields may already transition to high in West Africa, Southeast Asia and Central and South America (Faye et al. 2018⁴⁸) (medium confidence). For further information see Section 5.3.2.1. By contrast, higher latitudes may initially benefit from warming as well as well higher CO₂ concentrations (IPCC 2018a⁴⁹). Wheat yield losses are expected to be lower for the USA (−5.5 ± 4.4% per degree Celsius) and France (−6.0 ± 4.2% per degree Celsius) compared to India (−9.1 ± 5.4% per degree Celsius) (Zhao et al. 2017⁵⁰). Very high risks to low-latitude yields may occur between 3°C and 4°C (medium confidence). At these temperatures, catastrophic reductions in crop

Stability of food supply is expected to decrease as the magnitude and frequency of extreme events increase, disrupting food chains in all areas of the world (medium evidence, high agreement) (Wheeler and Von Braun 2013⁶⁰; Coates 2013⁶¹; Puma et al. 2015⁶²; Deryng et al. 2014⁶³; Harvey et al. 2014b⁶⁴; Iizumi et al. 2013⁶⁵; Seaman et al. 2014⁶⁶) (Sections 5.3.2, 5.3.3, 5.6.2 and 5.7.1). While international trade in food is assumed to be a key response for alleviating hunger, historical data and economic models suggest that international trade does not adequately redistribute food globally to offset yield declines or other food shortages when weather extremes reduce crop yields (medium confidence) (Schmitz et al. 2012⁶⁷; Chatzopoulos et al. 2019⁶⁸; Marchand et al. 2016⁶⁹; Gilbert 2010⁷⁰; Wellesley et al. 2017⁷¹). When droughts, heat waves, floods or other extremes destroy crops, evidence has shown that exports are constrained in key producing countries contributing to price spikes and social tension in importing countries which reduce access to food (medium evidence, medium agreement) (von Uexkull et al. 2016⁷²; Gleick 2014⁷³; Maystadt and Ecker 2014⁷⁴; Kelley et al. 2015⁷⁵; Church et al. 2017⁷⁶; Götz et al. 2013⁷⁷; Puma et al. 2015⁷⁸; Willenbockel 2012⁷⁹; Headey 2011⁸⁰; Distefano et al. 2018⁸¹; Brooks 2014⁸²). There is little understanding of how food system shocks cascade through a modern interconnected economy. Reliance on global markets may reduce some risks, but the ongoing globalisation of food trade networks exposes the world food system to new impacts that have not been seen in the past (Sections 5.1.2, 5.2.1, 5.5.2.5, 5.6.5 and 5.7.1). The global food system is vulnerable to systemic disruptions and increasingly interconnected inter-country food dependencies, and changes in the frequency and severity of extreme weather events may complicate future responses (Puma et al. 2015⁸³; Jones and Hiller 2017⁸⁴).

Impacts of climate change are already detectable on food supply and access as price and trade reactions have occurred in response to heatwaves, droughts and other extreme events (high evidence, high agreement) (Noble et al. 2014⁸⁵; O’Neill et al. 2017b⁸⁶). The impact of climate change on food stability is underexplored (Schleussner et al. 2016⁸⁷; James et al. 2017⁸⁸). However, some literature assesses that by about 2035, daily maximum temperatures will exceed the 90th percentile of historical (1961–1990) temperatures on 25–30% of days (O’Neill et al. 2017b⁸⁹, Figures 11–17) with negative shocks to food stability and world food prices. O’Neill et al. (2017b)⁹⁰ remark that in the future, return periods for precipitation events globally (land only) will reduce from one-in-20-year (historical) to about one-in-14- year or less by 2046–2065 in many areas of the world. Domestic efforts to insulate populations from food price spikes associated with climatic stressors in the mid-2000s have been shown to inadequately shield from poverty, and worsen poverty globally (Diffenbaugh et al. 2012⁹¹; Meyfroidt et al. 2013⁹²; Hertel et al. 2010⁹³). The transition to high risk is estimated to occur around 1.4°C, possibly by 2035, due to changes in temperature and heavy precipitation events (medium confidence) (O’Neill et al. 2017b⁹⁴; Fritsche et al. 2017a⁹⁵; Harvey et al. 2014b⁹⁶). Very high risk may occur by 2.4°C (medium confidence) and 4°C of warming is considered catastrophic (IPCC 2018c⁹⁷; Noble et al. 2014⁹⁸) for food stability and access because a combination of extreme events, compounding political and social factors, and shocks to crop yields can heavily constrain options to ensure food security in import- reliant countries.

Soil erosion increases risks of economic loss and declines in livelihoods due to reduced land productivity. In the EU, on-site costs of soil erosion by wind has been reported at an average of 55 USD per hectare annually, but up to 450 USD per hectare for sugar beet and oilseed rape (Middleton et al. 2017⁹⁹). Farmers in the Dapo watershed in Ethiopia lose about 220 USD per hectare of maize due to loss of nitrogen through soil erosion (Erkossa et al. 2015¹⁰⁰). Soil erosion not only increases crop loss but has been shown to have reduced household food supply with older farmers most vulnerable to losses from erosion (Ighodaro et al. 2016¹⁰¹). Erosion also results in increased risks to human health, through air pollution from aerosols (Middleton et al. 2017¹⁰²), and brings risks of reduced ES including supporting services related to soil formation.

At current levels of warming, changes in erosion are already detected in many regions. Attribution to climate change is challenging as there are other powerful drivers of erosion (e.g., land use), limited global- scale studies (Li and Fang 2016a¹⁰³; Vanmaercke et al. 2016a¹⁰⁴) and the absence of formal detection and attribution studies (Section 4.2.3). However, studies have found an increase in short-duration and high-intensity precipitation, due to anthropogenic climate change, which is a causative factor for soil erosion (Lenderink and van Meijgaard 2008¹⁰⁵; Li and Fang 2016b¹⁰⁶). High risks of erosion may occur between 2°C and 3.5°C (low confidence) as continued increases in intense precipitation are projected at these temperature thresholds (Fischer and Knutti 2015¹⁰⁷) in many regions. Warming also reduces soil organic matter, diminishing resistance against erosion. There is low confidence concerning the temperature threshold at which risks become very high due to large regional differences and limited global-scale studies (Li and Fang 2016b¹⁰⁸; Vanmaercke et al. 2016b¹⁰⁹) (Section 4.4).

Water scarcity in drylands contributes to changes in desertification and hazards such as dust storms, increasing risks of economic loss, declines in livelihoods of communities and negative health effects (high confidence) (Section 3.1.3). Further information specific to costs and impacts of water scarcity and droughts is detailed in Cross- Chapter Box 5 in Chapter 3.

The IPCC AR5 report and the SR15 concluded that there is low confidence in the direction of drought trends since 1950 at the global scale. While these reports did not assess water scarcity with a specific focus on drylands, they indicated that there is high confidence in observed drought increases in some regions of the world, including in the Mediterranean and West Africa (IPCC AR5) and that there is medium confidence that anthropogenic climate change has contributed to increased drying in the Mediterranean region (including southern Europe, northern Africa and the western Asia and the Middle east) and that this tendency will continue to increase under higher levels of global warming (IPCC 2018d). Some parts of the drylands have experienced decreasing precipitation over recent decades (IPCC AR5) (Chapter 3 and Section 3.2), consistent with the fact that climate change is implicated in desertification trends in some regions (Section 3.2.2). Dust storms, linked to changes in precipitation and vegetation, appear to be occurring with greater frequency in some deserts and their margins (Goudie 2014¹¹⁰) (Section 3.3.1). There is therefore high confidence that the transition from undetectable to moderate risk associated with water scarcity in drylands occurred in recent decades in the range 0.7°C to 1°C (Figure 7.1).

There are clear links between climate change and vegetation cover changes, tree mortality, forest diseases, insect outbreaks, forest fires, forest productivity and net ecosystem biome production (Allen et al. 2010¹¹⁵; Bentz et al. 2010¹¹⁶; Anderegg et al. 2013¹¹⁷; Hember et al. 2017¹¹⁸; Song et al. 2018¹¹⁹; Sturrock et al. 2011¹²⁰). Forest dieback, often a result of drought and temperature changes, not only produces risks to forest ecosystems but also to people with livelihoods dependent on forests. A 50-year study of temperate forest, dominated by beech (Fagus sylvatica L.), documented a 33% decline in basal area and a 70% decline in juvenile tree species, possibly as a result of interacting pressures of drought, overgrazing and pathogens (Martin et al. 2015¹²¹). There is high confidence that such dieback impacts ecosystem properties and services including soil microbial community structure (Gazol et al. 2018¹²²). Forest managers and users have reported negative emotional impacts from forest dieback such as pessimism about losses, hopelessness and fear (Oakes et al. 2016¹²³). Practices and policies such as forest classification systems, projection of growth, yield and models for timber supply are already being affected by climate change (Sturrock et al. 2011¹²⁴).

While risks to ecosystems and livelihoods from vegetation degradation are already detectable at current levels of GMT increase, risks are expected to reach high levels between 1.6°C and 2.6°C (medium confidence). Significant uncertainty exists due to countervailing factors: CO₂ fertilisation encourages forest expansion but increased drought, insect outbreaks, and fires result in dieback (Bonan 2008¹²⁵; Lindner et al. 2010¹²⁶). The combined effects of temperature and precipitation change, with CO₂ fertilisation, make future risks to forests very location specific. It is challenging therefore to make global estimates. However, even locally specific studies make clear that very high risks occur between 2.6°C and 4°C (medium confidence). Australian tropical rainforests experience significant loss of biodiversity with 3.5°C increase. At this level of increase there are no areas with greater than 30 species, and all endemics disappear from low- and mid-elevation regions (Williams et al. 2003¹²⁷). Mountain ecosystems are particularly vulnerable (Loarie et al. 2009¹²⁸).

Increasing fires result in heightened risks to infrastructure, accelerated erosion, altered hydrology, increased air pollution, and negative mental health impacts. Fire not only destroys property but induces changes in underlying site conditions (ground cover, soil water repellency, aggregate stability and surface roughness) which amplifies runoff and erosion, increasing future risks to property and human lives during extreme rainfall events (Pierson and Williams 2016¹²⁹). Dust and ash from fires can impact air quality in a wide area. For example, a dust plume from a fire in Idaho, USA, in September 2010 was visible in MODIS satellite imagery and extended at least 100 km downwind of the source area (Wagenbrenner et al. 2013¹³⁰). Individuals can suffer from property damage or direct injury, psychological trauma, depression, and post traumatic stress disorder, and have reported negative impacts to well-being from loss of connection to landscape (Paveglio et al. 2016¹³¹; Sharples et al. 2016a¹³²). Costs of large wildfires in the USA can exceed 20 million USD per day (Pierson et al. 2011¹³³) and has been estimated at 8.5 billion USD per year in Australia (Sharples et al. 2016b¹³⁴). Globally, human exposure to fire will increase due to projected population growth in fire-prone regions (Knorr et al. 2016a¹³⁵).

It is not clear how quickly, or even if, systems can recover from fires. Longevity of effects may differ depending on cover recruitment rate and soil conditions, recovering in one to two seasons or over 10 growing seasons (Pierson et al. 2011¹³⁶). In Russia, one-third of forest area affected by fires turned into unproductive areas where natural reforestation is not possible within 2–3 lifecycles of major forest forming species (i.e., 300–600 years) (Shvidenko et al. 2012¹³⁷).

Risks under current warming levels are already moderate as anthropogenic climate change has caused significant increases in fire area (high confidence) due to availability of detection and attribution studies) (Cross-Chapter Box 3 in Chapter 2). This has been detected and attributed regionally, notably in the western USA (Abatzoglou and Williams 2016¹³⁸; Westerling et al. 2006¹³⁹; Dennison et al. 2014¹⁵⁷³), Indonesia (Fernandes et al. 2017¹⁴⁰) and other regions (Jolly et al. 2015¹⁴¹). Regional increases have been observed despite a global- average declining trend induced by human fire-suppression strategies, especially in savannahs (Yang et al. 2014a¹⁴²; Andela et al. 2017¹⁴³).

There is a risk of damage to the natural and built environment from permafrost thaw-related ground instability. Residential, transportation, and industrial infrastructure in the pan-Arctic permafrost area are particularly at risk (Hjort et al. 2018¹⁵⁰). High risks already exist at low temperatures (high confidence). Approximately, 21–37% of Arctic permafrost is projected to thaw under a 1.5°C of warming (Hoegh-Guldberg et al. 2018¹⁵¹). This increases to very high risk around 2°C (between 1.8°C and 2.3°C) of temperature increase since pre-industrial times (medium confidence) with 35–47% of the Arctic permafrost thawing (Hoegh-Guldberg et al. 2018¹⁵²). If climate stabilised at 2°C, still approximately 40% of permafrost area would be lost (Chadburn et al. 2017¹⁵³), leading to nearly four million people and 70% of current infrastructure in the pan-Arctic permafrost area exposed to permafrost thaw and high hazard (Hjort et al. 2018¹⁵⁴). Indeed between 2°C and 3°C a collapse of permafrost may occur with a drastic biome shift from tundra to boreal forest (Drijfhout et al. 2015; SR15¹⁵⁵). There is mixed evidence of a tipping point in permafrost collapse, leading to enhanced greenhouse gas (GHG) emission – particularly methane – between 2°C and 3°C (Hoegh-Guldberg et al. 2018¹⁵⁶).

Socio-economic developments and policy choices that govern land–climate interactions are an important driver of risk, along with climate change (very high confidence). Risks under two different Shared Socio-economic Pathways (SSPs) were assessed using emerging literature. SSP1 is characterised by low population growth, reduced inequalities, land-use regulation, low meat consumption, and moderate trade (Riahi et al. 2017¹⁵⁷; Popp et al. 2017a¹⁵⁸). SSP3 is characterised by high population growth, higher inequalities, limited land-use regulation, resource-intensive consumption including meat-intensive diets, and constrained trade (for further details see Chapter 1 and Cross-Chapter Box 9 in Chapters 6 and 7). These two SSPs, among the set of five SSPs, were selected because they illustrate contrasting futures, ranging from low (SSP1) to high (SSP3) challenges to mitigation and adaptation. Figure 7.2 shows that for a given global mean temperature (GMT) change, risks are different under SSP1 compared to SSP3. In SSP1, global temperature change does not increase above 3°C even in the baseline case (i.e., with no additional mitigation measures) because in this pathway the combination of low population and autonomous improvements, for example, in terms of carbon intensity and/or energy intensity, effectively act as mitigation measures (Riahi et al. 2017¹⁵⁹). Thus Figure 7.2 does not indicate risks beyond this point in either SSP1 and SSP3. Literature based on such socio-economic and climate models is still emerging and there is a need for greater research on impacts of different pathways. There are few SSP studies exploring aspects of desertification and land degradation, but a greater number of SSP studies on food security (Supplementary Material). SSP1 reduces the vulnerability and exposure of human and natural systems and thus limits risks resulting from desertification, land degradation and food insecurity compared to SSP3 (high confidence).

Land-based adaptation relates to a particular category of adaptation measures relying on land management (Sanz et al. 2017¹⁸³). While most land-based adaptation options provide co-benefits for climate mitigation and other land challenges (Chapter 6 and Section 6.4.1), in some contexts adaptation measures can have adverse side effects, thus implying a risk to socio-ecological systems.

One example of risk is the possible decrease in farmer income when applying adaptive cropland management measures. For instance, conservation agriculture including the principle of no-till farming, contributes to soil erosion management (Chapter 6 and Section 6.2). Yet, no-till management can reduce crop yields in some regions, and although this effect is minimised when no-till farming is complemented by the other two principles of conservation agriculture (residue retention and crop rotation), this could induce a risk to livelihood in vulnerable smallholder farming systems (Pittelkow et al. 2015¹⁸⁴).

Another example is the use of irrigation against water scarcity and drought. During the long lasting drought from 2007–2009 in California, USA, farmers adapted by relying on groundwater withdrawal and caused groundwater depletion at unsustainable levels (Christian-Smith et al. 2015¹⁸⁵). The long-term effects of irrigation from groundwater may cause groundwater depletion, land subsidence, aquifer overdraft, and saltwater intrusion (Tularam and Krishna 2009¹⁸⁶). Therefore, it is expected to increase the vulnerability of coastal aquifers to climate change due to groundwater usage (Ferguson and Gleeson 2012¹⁸⁷). The long-term practice of irrigation from groundwater may cause a severe combination of potential side effects and consequently irreversible results.

While historically land-use activities have been a net source of GHG emissions, in future decades the land sector will not only need to reduce its emissions, but also to deliver negative emissions through carbon dioxide removal (CDR) to reach the objective of limiting global warming to 2°C or below (Section 2.5).Although land-based mitigation in itself is a risk-reduction strategy aiming at abating climate change, it also entails risks to humans and ecosystems, depending on the type of measures and the scale of deployment. These risks fall broadly into two categories: risk of mitigation failure – due to uncertainties about mitigation potential, potential for sink reversal and moral hazard; and risks arising from adverse side effects – due to increased competition for land and water resources. This section focuses specifically on bioenergy and bioenergy with carbon capture and storage (BECCS) since it is one of the most prominent land-based mitigation strategies in future mitigation scenarios (along with large-scale forest expansion, which is discussed in Cross-Chapter Box 1 in Chapter 1). Bioenergy and BECCS is assessed in Chapter 6 as being, at large scales, the only response option with adverse side effects across all dimensions (adaptation, food security, land degradation and desertification) (Section 6.4.1).

Risk of mitigation failure. The mitigation potential from bioenergy and BECCS is highly uncertain, with estimates ranging from 0.4 to 11.3 GtCO₂e yr^–1 for the technical potential, while consideration of sustainability constraints suggest an upper end around 5 GtCO²e yr^–1 (Chapter 2, Section 2.6). In comparison, IAM-based mitigation pathways compatible with limiting global warming at 1.5°C project bioenergy and BECCS deployment exceeding this range (Figure 2.24 in Chapter 2). There is medium confidence that IAMs currently do not reflect the lower end and exceed the upper end of bioenergy and BECCS mitigation potential estimates (Anderson and Peters 2016¹⁸⁸; Krause et al. 2018¹⁸⁹; IPCC 2018c¹⁹⁰), with implications for the risk associated with reliance on bioenergy and BECCS deployment for climate mitigation.

In addition, land-based CDR strategies are subject to a risk of carbon sink reversal. This implies a fundamental asymmetry between mitigation achieved through fossil fuel emissions reduction compared to CDR. While carbon in fossil fuel reserves – in the case of avoided fossil fuel emissions – is locked permanently (at least over a time scale of several thousand years), carbon sequestered into the terrestrial biosphere – to compensate fossil fuel emissions – is subject to various disturbances, in particular from climate change and associated extreme events (Fuss et al. 2018¹⁹¹; Dooley and Kartha 2018¹⁹²). The probability of sink reversal therefore increases with climate change, implying that the effectiveness of land-based mitigation depends on emission reductions in other sectors and can be sensitive to temperature overshoot (high confidence). In the case of bioenergy associated with CCS (BECCS), the issue of the long-term stability of the carbon storage is linked to technical and geological constraints, independent of climate change but presenting risks due to limited knowledge and experience (Chapter 6 and Cross-Chapter Box 7 in Chapter 6).

Another factor in the risk of mitigation failure, is the moral hazard associated with CDR technologies. There is medium evidence and medium agreement that the promise of future CDR deployment – bioenergy and BECCS in particular – can deter or delay ambitious emission reductions in other sectors (Anderson and Peters 2016¹⁹³; Markusson et al. 2018a¹⁹⁴; Shue 2018a¹⁹⁵). The consequences are an increased pressure on land with higher risk of mitigation failure and of temperature overshoot, and a transfer of the burden of mitigation and unabated climate change to future generations. Overall, there is therefore medium evidence and high agreement that prioritising early decarbonisation with minimal reliance on CDR decreases the risk of mitigation failure and increases intergenerational equity (Geden et al. 2019¹⁹⁶; Larkin et al. 2018¹⁹⁷; Markusson et al. 2018b¹⁹⁸; Shue 2018b¹⁹⁹).

There is very high confidence that all the risk transitions occur at lower bioenergy levels in SSP3, implying higher risks associated with bioenergy deployment, due to the higher competition for land in this pathway. In SSP3, land-based mitigation is therefore strongly limited by sustainability constraints such that moderate risk occur already between 0.5 and 1.5 million km² (medium confidence). There is medium confidence that a bioenergy footprint beyond 4 to 8 million km² would entail very high risk with transgression of most planetary boundaries (Heck et al. 2018b²⁰⁸), strong decline in sustainability indicators (Humpenöder et al. 2017²⁰⁹) and increase in the population at risk of hunger well above 100 million (Fujimori et al. 2018a²¹⁰; Hasegawa et al. 2018b²¹¹).

Windows of opportunity are important learning moments wherein an event or disturbance in relation to land, climate, and food security triggers responsive social, political, policy change (medium agreement). Policies play an important role in windows of opportunity and are important in relation to managing risks of desertification, soil degradation, food insecurity, and supporting response options for SLM (high agreement) (Kivimaa and Kern 2016³³⁷; Gupta et al. 2013b³³⁸; Cosens et al. 2017³³⁹; Darnhofer 2014³⁴⁰; Duru et al. 2015³⁴¹) (Chapter 6).

A wide range of events or disturbances may initiate windows of opportunity – ranging from climatic events and disasters, recognition of a state of land degradation, an ecological social or political crisis, and a triggered regulatory burden or opportunity. Recognition of a degraded system such as land degradation and desertification (Chapters 3 and 4) and associated ecosystem feedbacks, allows for strategies, response options and policies to address the degraded state (Nyström et al. 2012³⁴³). Climate related disasters (flood, droughts, etc.) and crisis may trigger latent local adaptive capacities leading to systemic equitable improvement (McSweeney and Coomes 2011³⁴⁴), or novel and innovative recombining of sources of experience and knowledge, allowing navigation to transformative social ecological transitions (Folke et al. 2010³⁴⁵). The occurrence of a series of punctuated crises such as floods or droughts, qualify as windows of opportunity when they enhance society’s capacity to adapt over the long term (Pahl-Wostl et al. 2013³⁴⁶). A disturbance from an ecological, social, or political crisis may be sufficient to trigger the emergence of new approaches to governance wherein there is a change in the rules of the social world such as informal agreements surrounding human activities or formal rules of public policies (Olsson et al. 2006³⁴⁷; Biggs et al. 2017³⁴⁸) (Section 7.6). A combination of socio-ecological changes may provide windows of opportunity for a socio-technical niche to be adopted on a greater scale, transforming practices towards SLM such as biodiversity-based agriculture (Darnhofer 2014³⁴⁹; Duru et al. 2015³⁵⁰).

Food security is affected by interactions between climatic factors (rising temperatures, changes in weather variability and extremes), changes in land use and land degradation, and Socio-economic Pathways and policy choices related to food systems (see Figures 7.1 and 7.2). As outlined in Chapter 5, key aspects of food security are food availability, access to food, utilisation of food, and stability of food systems.

While comprehensive reviews of policy are rare and additional data is needed (Adu et al. 2018³⁶⁷), evidence indicates that the results of food security interventions vary widely due to differing values underlying the design of instruments. A large portfolio of measures is available to shape outcomes in these areas from the use of tariffs or subsidies, to payments for production practices (OECD 2018³⁶⁸). In the past, efforts to increase food production through significant investment in agricultural research, including crop improvement, have benefited farmers by increasing yields and reducing losses, and have helped consumers by lowering food prices (Pingali 2012 ¹⁶⁷⁷, 2015¹⁶⁷⁸; Alston and Pardey 2014³⁶⁹; Popp et al. 2013³⁷⁰). Public spending on agriculture research and development (R&D) has been more effective at raising sustainable agriculture productivity than irrigation or fertiliser subsidies (OECD 2018³⁷¹). Yet, on average, between 2015 and 2017, governments spent only around 14% of total agricultural support on services, including physical and knowledge infrastructure, transport and information and communications technology.

In terms of increasing food availability and supply, producer support, including policies mandating subsidies or payments, have been used to boost production of certain commodities or protect ES. Incentives can distort markets and farm business decisions in both negative and positive ways. For example, the European Union promotes meat and dairy production through voluntary coupled direct payments. These do not yet internalise external damage to climate, health, and groundwater (Velthof et al. 2014³⁷²; Bryngelsson et al. 2016³⁷³). In most countries, producer support has been declining since the mid-1990s (OECD 2018³⁷⁴). Yet new evidence indicates that a government policy supporting producer subsidy could encourage farmers to adopt new technologies and reduce GHG emissions in agriculture (medium evidence, high agreement). However, this will require large capital (Henderson 2018³⁷⁵). Since a 1995 reform in its forest law, Costa Rica has effectively used a combination of fuel tax, water tax, loans and agreements with companies, to pay landowners for agroforestry, reforestation and sustainable forest management (Porras and Asquith 2018³⁷⁶).

Inland capture fisheries and aquaculture are an integral part of nutrition security and livelihoods for large numbers of people globally (Welcomme et al. 2010³⁷⁷; Hall et al. 2013³⁷⁸; Tidwell and Allan 2001³⁷⁹; Youn et al. 2014³⁸⁰) and are increasingly vulnerable to climate change and competing land and water use (Allison et al. 2009³⁸¹; Youn et al. 2014³⁸²). Future production may increase in some high-latitude regions (low confidence) but production is likely to decline in low-latitude regions under future warming (high confidence) (Brander and Keith 2015³⁸³; Brander 2007³⁸⁴). However over-exploitation and degradation of rivers has resulted in a decreasing trend in the contribution of capture fisheries to protein security in comparison to managed aquaculture (Welcomme et al. 2010³⁸⁵). Aquaculture, however, competes for land and water resources with many negative ecological and environmental impacts (Verdegem and Bosma 2009³⁸⁶; Tidwell and Allan 2001³⁸⁷). Inland capture fisheries are undervalued in national and regional food security, ES and economy, are data deficient and are neglected in terms of supportive policies at national levels, and absent in SDGs (Cooke et al. 2016³⁸⁸; Hall et al. 2013³⁸⁹; Lynch et al. 2016³⁹⁰). Revival of sustainable capture fisheries and converting aquaculture to environmentally less-damaging management regimes, is likely to succeed with the following measures: investment in recognition of their importance, improved valuation and assessment, secure tenure and adoption of social, ecological and technological guidelines, upstream-downstream river basin cooperation, and maintenance of ecological flow regimes in rivers (Youn et al. 2014³⁹¹; Mostert et al. 2007³⁹²; Ziv et al. 2012³⁹³; Hurlbert and Gupta 2016³⁹⁴; Poff et al. 2003³⁹⁵; Thomas 1996³⁹⁶; FAO 2015a³⁹⁷).

Extension services, and policies supporting agricultural extension systems, are also critical. Smallholder farmer-dominated agriculture is currently the backbone of global food security in the developing world. Without education and incentives to manage land and forest resources in a manner that allows regeneration of both the soils and wood stocks, smallholder farmers tend to generate income through inappropriate land management practices, engage in agricultural production on unsuitable land and use fertile soils, timber and firewood for brick production and construction. Also, they engage in charcoal production (deforestation) as a coping mechanism (increasing income) against food deficiency (Munthali and Murayama 2013³⁹⁸). Through extension services, governments can play a proactive role in providing information on climate and market risks, animal and plant health. Farmers with greater access to extension training retain more crop residues for mulch on their fields (Jaleta et al. 2015¹⁶⁷⁹, 2013¹⁶⁸⁰; Baudron et al. 2014³⁹⁹).

Food security cannot be achieved by increasing food availability alone. Policy instruments, which increase access to food at the household level, include safety-net programming and universal basic income. The graduation approach, developed and tested over the past decade using randomised control trials in six countries, has lasting positive impacts on income, as well as food and nutrition security (Banerjee et al. 2015⁴⁰⁰; Raza and Poel 2016⁴⁰¹) (robust evidence, high agreement). The graduation approach layers and integrates a series of interventions designed to help the poorest: consumption support in the form of cash or food assistance, transfer of an income- generating asset (such as a livestock) and training on how to maintain the asset, assistance with savings and coaching or mentoring over a period of time to reinforce learning and provide support. Due to its success, the graduation approach is now being scaled up, and is now used in more than 38 countries and included by an increasing number of governments in social safety-net programmes (Hashemi and de Montesquiou 2011⁴⁰²).

There is medium evidence and high agreement from all regions of the world that safety nets and social protection schemes can provide stability which prevents and reduces abject poverty (Barrientos 2011⁴¹⁹; Hossain 2018⁴²⁰; Cook and Pincus 2015⁴²¹; Huang and Yang 2017⁴²²; Slater 2011⁴²³; Sparrow et al. 2013⁴²⁴; Rodriguez-Takeuchi and Imai 2013⁴²⁵; Bamberg et al. 2018⁴²⁶) in the face of climatic stressors and land change (Davies et al. 2013⁴²⁷; Cutter et al. 2012b⁴²⁸; Pelling 2011⁴²⁹; Ensor 2011⁴³⁰).

The World Bank estimates that, globally, social safety net transfers have reduced the absolute poverty gap by 45% and the relative poverty gap by 16% (World Bank 2018⁴³¹). Adaptive social protection builds household capacity to deal with shocks as well as the capacity of social safety nets to respond to shocks. For low-income communities reliant on land and climate for their livelihoods and well-being, social protection provides a way for vulnerable groups to manage weather and climatic variability and deteriorating land conditions to household income and assets (robust evidence, high agreement) (Baulch et al. 2006⁴³²; Barrientos 2011⁴³³; Harris 2013⁴³⁴; Fiszbein et al. 2014⁴³⁵; Kiendrebeogo et al. 2017⁴³⁶; Kabeer et al. 2010⁴³⁷; FAO 2015b⁴³⁸; Warner et al. 2018⁴³⁹; World Bank 2018⁴⁴⁰).

A lifecycle approach to social protection is one approach, which some countries (such as Bangladesh) are using when developing national social protection policies. These policies acknowledge that households face risks across the lifecycle that they need to be protected from. If shocks are persistent, or occur numerous times, then policies can address concerns of a more structural nature (Glauben et al. 2012⁴⁴¹). Barrett (2005)⁴⁴², for example, distinguishes between the role of safety nets (which include programmes such as emergency feeding programmes, crop or unemployment insurance, disaster assistance, etc.) and cargo nets (which include land reforms, targeted microfinance, targeted school food programmes, etc.). While the former prevents non-poor and transient poor from becoming chronically poor, the latter is meant to lift people out of poverty by changing societal or institutional structures. The graduation approach has adopted such systematic thinking with successful results (Banerjee et al. 2015⁴⁴³).

Social protection systems can provide buffers against shocks through vertical or horizontal expansion, ‘piggybacking’ on pre-established programmes, aligning social protection and humanitarian systems or refocusing existing resources (Wilkinson et al. 2018⁴⁴⁴; O’Brien et al. 2018⁴⁴⁵; Jones and Presler-Marshall 2015⁴⁴⁶). There is increasing evidence that forecast-based financing, linked to a social protection, can be used to enable anticipatory actions based on forecast triggers, and guarantee funding ahead of a shock (Jjemba et al. 2018⁴⁴⁷). Accordingly, scaling up social protection based on an early warning could enhance timeliness, predictability and adequacy of social protection benefits (Kuriakose et al. 2012⁴⁴⁸; Costella et al. 2017a⁴⁴⁹; Wilkinson et al. 2018⁴⁵⁰; O’Brien et al. 2018⁴⁵¹).

Risk management addressing climate change has broadened to include mitigation, adaptation and disaster preparedness in a process using instruments facilitating contingency and cross-sectoral planning (Hurlimann and March 2012⁴⁸¹; Oels 2013⁴⁸²), social community planning, and strategic, long-term planning (Serrao-Neumann et al. 2015a⁴⁸³). A comprehensive consideration integrates principles from informal support mechanisms to enhance formal social protection programming (Mobarak and Rosenzweig 2013⁴⁸⁴; Stavropoulou et al. 2017⁴⁸⁵) such that the social safety net, disaster risk management, and climate change adaptation are all considered to enhance livelihoods of the chronic poor (see char dwellers and recurrent floods in Jamuna and Brahmaputra basins of Bangladesh Awal 2013) (Section 7.4.7). Iterative risk management is an ongoing process of assessment, action, reassessment and response (Mochizuki et al. 2015⁴⁸⁷) (Sections 7.5.2 and 7.4.7.2).

Important elements of risk planning include education, and creation of hazard and risk maps. Important elements of predicting include hydrological and meteorological monitoring to forecast weather, seasonal climate forecasts, aridity, flood and extreme weather. Effective responding requires robust communication systems that pass on information to enable response (Cools et al. 2016⁴⁸⁸).

Gauging the effectiveness of policy instruments is challenging. Timescales may influence outcomes. To evaluate effectiveness researchers, programme managers and communities strive to develop consistency, comparability, comprehensiveness and coherence in their tracking. In other words, practitioners utilise a consistent and operational conceptualisation of adaptation; focus on comparable units of analysis; develop comprehensive datasets on adaptation action; and are coherent with an understanding of what constitutes real adaptation (Ford and Berrang-Ford 2016⁴⁸⁹). Increasing the use of systematic reviews or randomised evaluations may also be helpful (Alverson and Zommers 2018⁴⁹⁰).

Many risk management policy instruments are referred to by the International Organization of Standardization which lists risk management principles, guidelines, and frameworks for explaining the elements of an effective risk management programme (ISO 2009⁴⁹¹). The standard provides practical risk management instruments and makes a business case for risk management investments (McClean et al. 2010⁴⁹²). Insurance addresses impacts associated with extreme weather events (storms, floods, droughts, temperature extremes), but it can provide disincentives for reducing disaster risk at the local level through the transfer of risk spatially to other places or temporally to the future (Cutter et al. 2012b⁴⁹³) and uptake is unequally distributed across regions and hazards (Lal et al. 2012⁴⁹⁴). Insurance instruments (Sections 7.4.2 and 7.4.6) can take many forms (traditional indemnity based, market-based crop insurance, property insurance), and some are linked to livelihoods sensitive to weather as well as food security (linked to social safety-net programmes) and ecosystems (coral reefs and mangroves). Insurance instruments can also provide a framework for risk signals to adaptation planning and implementation and facilitate financial buffering when climate impacts exceed current capabilities delivered through both public and private finance (Bogale 2015b⁴⁹⁵; Greatrex et al. 2015⁴⁹⁶; Surminski et al. 2016⁴⁹⁷). A holistic consideration of all instruments responding to extreme impacts of climate change (drought, flood, etc.) is required when assessing if policy instruments are promoting livelihood capitals and contributing to the resilience of people and communities (Hurlbert 2018b⁴⁹⁸). This holistic consideration of policy instruments leads to a consideration of risk governance (Section 7.6).

A more detailed review of drought instruments, and three broad policy approaches for responding to drought, is provided in Cross- Chapter Box 5 in Chapter 3. Three broad approaches include: (i) early warning systems and response to the disaster of drought (through instruments such as disaster assistance or crop insurance); (ii) disaster response ex-ante preparation (through drought preparedness plans); and (iii) drought risk mitigation (proactive polices to improve water-use efficiency, make adjustments to water allocation, funds or loans to build technology such as dugouts or improved soil management practices).

Drought plans are still predominantly reactive crisis management plans rather than proactive risk management and reduction plans. Reactive crisis management plans treat only the symptoms and are inefficient drought management practices. More efficient drought preparedness instruments are those that address the underlying vulnerability associated with the impacts of drought, thereby building agricultural producer adaptive capacity and resilience (high confidence) (Cross-Chapter Box 5 in Chapter 3).

There is robust evidence and high agreement that fire strategies need to be tailored to site-specific conditions in an adaptive application that is assessed and reassessed over time (Dellasala et al. 2004⁵¹⁶; Rocca et al. 2014⁵¹⁷). Strategies for fire management include fire suppression, prescribed fire and mechanical treatments (such as thinning the canopy), and allowing wildfire with little or no active management (Rocca et al. 2014⁵¹⁸). Fire suppression can degrade the effectiveness of forest fire management in the long run (Collins et al. 2013⁵¹⁹).

Different forest types have different fire regimes and require different fire management policies (Dellasala et al. 2004⁵²⁰). For instance, Cerrado, a fire dependent savannah, utilises a different fire management policy and fire suppression policy (Durigan and Ratter 2016⁵²¹). The choice of strategy depends on local considerations, including land ownership patterns, dynamics of local meteorology, budgets, logistics, federal and local policies, tolerance for risk and landscape contexts. In addition, there are trade-offs among the management alternatives and often no single management strategy will simultaneously optimise ES, including water quality and quantity, carbon sequestration, or run- off erosion prevention (Rocca et al. 2014⁵²²).

Flood risk management consists of command and control measures, including spatial planning and engineered flood defences (Filatova 2014⁵²³), financial incentive instruments issued by regional or national governments to facilitate cooperative approaches through local planning, enhancing community understanding and political support for safe development patterns and building standards, and regulations requiring local government participation and support for local flood planning (Burby and May 2009⁵²⁴). However, Filatova (2014) found that if autonomous adaptation is downplayed, people are more likely to make land-use choices that collectively lead to increased flood risks and leave costs to governments. Taxes and subsidies that do not encourage (and even counter) perverse behaviour (such as rebuilding in flood zones) are important instruments mitigating this cost to government. Flood insurance has been found to be maladaptive as it encourages rebuilding in flood zones (O’Hare et al. 2016⁵²⁵) and government flood disaster assistance negatively impacts on average insurance coverage the following year (Kousky et al. 2018a⁵²⁶). Modifications to flood insurance can counter perverse behaviour. One example is the provision of discounts on flood insurance for localities that undertake one of 18 flood mitigation activities, including structural mitigation (constructing dykes, dams, flood control reservoirs), and non-structural initiatives such as point source control and watershed management efforts, education and maintenance of flood-related databases (Zahran et al. 2010⁵²⁷). Flood insurance that provides incentives for flood mitigation, marketable permits and transferable development rights (see Case study: Flood and food security in Section 7.6) instruments can provide price signals to stimulate autonomous adaptation, countering barriers of path dependency, and the time lag between private investment decisions and consequences (Filatova 2014⁵²⁸). To build adaptive capacity, consideration needs to be made of policy instruments responding to flood, including flood zone mapping, land-use planning, flood zone building restrictions, business and crop insurance, disaster assistance payments, preventative instruments, (including environmental farm planning, e.g., soil and water management (see Chapter 6)), farm infrastructure projects, and recovery from debilitating flood losses ultimately through bankruptcy (Hurlbert 2018a⁵²⁹). Non-structural measures have been found to advance sustainable development as they are more reversible, commonly acceptable and environmentally friendly (Kundzewicz 2002⁵³⁰).

Pathways reflecting current nationally stated mitigation ambitions as submitted under the Paris Agreement would not limit global warming to 1.5°C with no or limited overshoot, but instead result in a global warming of about 3°C by 2100 with warming continuing afterward (IPCC 2018d). Reversing warming after an overshoot of 0.2°C or higher during this century would require deployment of CDR at rates and volumes that might not be achievable given considerable implementation challenges (IPCC 2018d). This gap (Höhne et al. 2017⁵³¹; Rogelj et al. 2016⁵³²) creates a significant risk of global warming impacting on land degradation, desertification, and food security (IPCC 2018d⁵³³) (Section 7.2). Action can be taken by 2030 adopting already known cost-effective technology (United Nations Environment Programme 2017⁵³⁴), improving the finance, capacity building, and technology transfer mechanisms of the United Nations Framework Convention on Climate Change (UNFCCC), improving food security (listed by 73 nations in their nationally determined contributions (NDCs)) and nutritional security (listed by 25 nations) (Richards et al. 2015⁵³⁵). UNFCCC Decision 1. CP21 reaffirmed the UNFCCC target that ‘developed country parties provide USD 100 billion annually by 2020 for climate action in developing countries’ (Rajamani 2011⁵³⁶) and a new collective quantified goal above this floor is to be set, taking into account the needs and priorities of developing countries (Fridahl and Linnér 2016⁵³⁷).

Mitigation policy instruments to address this shortfall include financing mechanisms, carbon pricing, cap and trade or emissions trading, and technology transfer. While climate change is a global commons problem containing free-riding issues cost-effective international policies that ensure that countries get the most environmental benefit out of mitigation investments promote an international climate policy regime (Nordhaus 1999⁵³⁸; Aldy and Stavins 2012⁵³⁹). Carbon pricing instruments may provide an entry point for inclusion of appropriate agricultural carbon instruments. Models of cost-efficient distribution of mitigation across regions and sectors typically employ a global uniform carbon price, but such treatment in the agricultural sector may impact on food security (Section 7.4.4.4).

One policy initiative to advance climate mitigation policy coherence in this section is the phase out of subsidies for fossil fuel production (see also Section 7.4.8). The G20 agreed in 2009, and the G7 agreed in 2016, to phase out these subsidies by 2025. Subsidies include lower tax rates or exemptions and rebates of taxes on fuels used by particular consumers (diesel fuel used by farming, fishing, etc.), types of fuel, or how fuels are used. The OECD estimates the overall value of these subsides to be 160–200 billion USD annually between 2010 and 2014 (OECD 2015⁵⁴⁰). The phase-out of fossil fuel subsidies has important economic, environmental and social benefits. Coady et al. (2017)⁵⁴¹ estimate the economic and environmental benefits of reforming fossil fuel subsidies could be valued worldwide at 4.9 trillion USD in 2013, and 5.3 trillion USD in 2015. Eliminating subsidies could have reduced emissions by 21%, raised 4% of global GDP as revenue (in 2013), and improved social welfare (Coady et al. 2017⁵⁴²).

Legal instruments addressing perceived deficiencies in climate change mitigation include human rights and liability. Developments in attribution science are improving the ability to detect human influence on extreme weather. Marjanac et al. (2017)⁵⁴³ argue that this broadens the legal duty of government, business and others to manage foreseeable harms, and may lead to more climate change litigation (Marjanac et al. 2017)⁵⁴⁴. Peel and Osofsky (2017)⁵⁴⁵ argue that courts are becoming increasingly receptive to employ human rights claims in climate change lawsuits (Peel and Osofsky 2017⁵⁴⁶); citizen suits in domestic courts are not a universal phenomenon and, even if unsuccessful, Estrin (2016)⁵⁴⁷ concludes they are important in underlining the high level of public concern.

Similar instruments for mitigation could be applied to the land sector as in other sectors, including: market-based measures such as taxes and cap and trade systems; standards and regulations; subsidies and tax credits; information instruments and management tools; R&D investment; and voluntary compliance programmes. However, few regions have implemented agricultural mitigation instruments Cooper et al. 2013⁵⁴⁸). Existing regimes focus on subsidies, grants and incentives, and voluntary offset programmes.

Although carbon pricing is recognised to be an important cost- effective instrument in a portfolio of climate policies (high evidence, high agreement) (Aldy et al. 2010⁵⁴⁹), as yet, no country is exposing their agricultural sector emissions to carbon pricing in any comprehensive way. A carbon tax, fuel tax, and carbon markets (cap and trade system or Emissions Trading System (ETS), or baseline and credit schemes, and voluntary markets) are predominant policy instruments that implement carbon pricing. The advantage of carbon pricing is environmental effectiveness at relatively low cost (high evidence, high agreement) (Baranzini et al. 2017⁵⁵⁰; Fawcett et al. 2014⁵⁵¹). Furthermore, carbon pricing could be used to raise revenue to reinvest in public spending, either to help certain sectors transition to lower carbon systems, or to invest in public spending unrelated to climate change. Both of these options may make climate policies more attractive and enhance overall welfare (Siegmeier et al. 2018⁵⁵²), but there is, as yet, no evidence of the effectiveness of emissions pricing in agriculture (Grosjean et al. 2018⁵⁵³). There is, however, a clear need for progress in this area as, without effective carbon pricing, the mitigation potential identified in chapters 5 and 6 of this report will not be realised (high evidence, high agreement) (Boyce 2018⁵⁵⁴).

The price may be set at the social cost of carbon (the incremental impact of emitting an additional tonne of CO₂, or the benefit of slightly reducing emissions), but estimates of the SCC vary widely and are contested (high evidence, high agreement) (Pezzey 2019⁵⁵⁵). An alternative to the SCC includes a pathways approach that sets an emissions target and estimates the carbon prices required to achieve this at the lowest possible cost (Pezzey 2019⁵⁵⁶). Theoretically, higher costs throughout the entire economy result in reduction of carbon intensity, as consumers and producers adjust their decisions in relation to prices corrected to reflect the climate externality (Baranzini et al. 2017⁵⁵⁷).

Both carbon taxes and cap and trade systems can reduce emissions, but cap and trade systems are generally more cost effective (medium evidence, high agreement) (Haites 2018a⁵⁵⁸). In both cases, the design of the system is critical to its effectiveness at reducing emissions (high evidence, high agreement) (Bruvoll and Larsen 2004⁵⁵⁹; (Lin and Li 2011⁵⁶⁰). The trading system allows the achievement of emission reductions in the most cost-effective manner possible and results in a market and price on emissions that create incentives for the reduction of carbon pollution. The way allowances are allocated in a cap and trade system is critical to its effectiveness and equity. Free allocations can be provided to trade-exposed sectors, such as agriculture, either through historic or output-based allocations, the choice of which has important implications (Quirion 2009⁵⁶¹). Output-based allocations may be most suitable for agriculture, also minimising leakage risk (see below in this section) (Grosjean et al. 2018⁵⁶²; Quirion 2009⁵⁶³). There is medium evidence and high agreement that properly designed, a cap and trade system can be a powerful policy instrument (Wagner 2013⁵⁶⁴) and may collect more rents than a variable carbon tax (Siegmeier et al. 2018⁵⁶⁵; Schmalensee and Stavins 2017⁵⁶⁶).

In the land sector, carbon markets are challenging to implement. Although several countries and regions have an ETS in place (for example, the EU, Switzerland, the Republic of Korea, Quebec in Canada, California in the USA (Narassimhan et al. 2018⁵⁶⁷)), none have included non-CO₂ (methane and nitrous oxide) emissions from agriculture. New Zealand is the only country currently considering ways to incorporate agriculture into its ETS (see Case study: Including agriculture in the New Zealand Emissions Trading Scheme).

Three main reasons explain the lack of implementation to date:

1. The large number of heterogeneous buyers and sellers, combined with the difficulties of monitoring, reporting and verification (MRV) of emissions from biological systems introduce potentially high levels of complexity (and transaction costs). Effective policies therefore depend on advanced MRV systems which are lacking in many (particularly developing) countries (Wilkes et al. 2017)⁵⁶⁸. This is discussed in more detail in the case study on the New Zealand Emissions Trading Scheme.
2. Adverse distributional consequences (Grosjean et al. 2018⁵⁶⁹) (medium evidence, high agreement). Distributional issues depend, in part, on the extent that policy costs can be passed on to consumers, and there is medium evidence and medium agreement that social equity can be increased through a combination of non-market and market-based instruments (Haites 2018b⁵⁷⁰).
3. Regulation, market-based or otherwise, adopted in only one jurisdiction and not elsewhere may result in ‘leakage’ or reduced effectiveness – where production relocates to weaker regulated regions, potentially reducing the overall environmental benefit. Although modelling studies indicate the possibility of leakage following unilateral agricultural mitigation policy implementation (e.g., Fellmann et al. 2018), there is no empirical evidence from the agricultural sector yet available. Analysis from other sectors shows an overestimation of the extent of carbon leakage in modelling studies conducted before policy implementation compared to evidence after the policy was implemented (Branger and Quirion 2014⁵⁷¹). Options to avoid leakage include: border adjustments (emissions in non-regulated imports are taxed at the border, and payments made on products exported to non-regulated countries are rebated); differential pricing for trade-exposed products; and output-based allocation (which effectively works as a subsidy for trade-exposed products). Modelling shows that border adjustments are the most effective at reducing leakage, but may exacerbate regional inequality (Böhringer et al. 2012⁵⁷²) and through their trade-distorting nature may contravene World Trade Organization rules. The opportunity for leakage would be significantly reduced, ideally through multi- lateral commitments (Fellmann et al. 2018⁵⁷³) (medium evidence, high agreement) but could also be reduced through regional or bi-lateral commitments within trade agreements.

Technology transfer has been part of the UNFCCC process since its inception and is a key element of international climate mitigation and adaptation efforts under the Paris Agreement. The IPCC definition of ‘technology transfer’ includes transfer of knowledge and technological cooperation (see Glossary) and can include modifications to suit local conditions and/or integration with indigenous technologies (Metz et al. 2000¹⁶⁷⁶). This definition suggests greater heterogeneity in the applications for climate mitigation and adaptation, especially in land-use sectors where indigenous knowledge may be important for long-term climate resilience (Nyong et al. 2007⁵⁷⁴). For land-use sectors, the typical reliance on trade and patent data for empirical analyses is generally not feasible as the ‘technology’ in question is often related to resource management and is neither patentable nor tradable (Glachant and Dechezleprêtre 2017⁵⁷⁵) and ill-suited to provide socially beneficially innovation for poorer farmers in developing countries (Lybbert and Sumner 2012⁵⁷⁶; Baker et al. 2017⁵⁷⁷).

Technology transfer has contributed to emissions reductions (medium confidence). A detailed study for nearly 4000 Clean Development Mechanism (CDM) projects showed that 39% of projects had a stated and actual technology transfer component, accounting for 59% of emissions reductions; however, the more land-intensive projects (e.g., afforestation, bioenergy) showed lower percentages (Murphy et al. 2015⁵⁷⁸). Bioenergy projects that rely on agricultural residues offer substantially more development benefits than those based on industrial residues from forests (Lee and Lazarus 2013⁵⁷⁹). Energy projects tended to have a greater degree of technology transfer under the CDM compared to non-energy projects (Gandenberger et al. 2016⁵⁸⁰). However, longer-term cooperation and collaborative R&D approaches to technology transfer will be more important in land-use sectors (compared to energy or industry) due to the time needed for improved resource management and interaction between researchers, practitioners and policymakers. These approaches offer longer-term technology transfer that is more difficult to measure compared to specific cooperation projects; empirical research on the effects of R&D collaboration could help to avoid the ‘one-policy-fits- all’ approach (Ockwell et al. 2015⁵⁸¹).

There is increasing recognition of the role of technology transfer in climate adaptation, but in the land-use sector there are inherent adoption challenges specific to adaptation, due to uncertainties arising from changing climatic conditions, agricultural prices, and suitability under future conditions (Biagini et al. 2014⁵⁸²). Engaging the private sector is important, as adoption of new technologies can only be replicated with significant private sector involvement (Biagini and Miller 2013⁵⁸³).

New cooperative mechanisms under the Paris Agreement illustrate the shift away from the Kyoto Protocol’s emphasis on obligations of developed country Parties to pursue investments and technology transfer, to a more pragmatic, decentralised and collaborative approach (Savaresi 2016⁵⁸⁴; Jiang et al. 2017⁵⁸⁵). These approaches can effectively include any combination of measures or instruments related to adaptation, mitigation, finance, technology transfer and capacity building, which could be of particular interest in land-use sectors where such aspects are more intertwined than in energy or industry sectors. Article 6 sets out several options for international cooperation (Gupta and Dube 2018⁵⁸⁶).

The close relationship between emission reductions, adaptive capacity, food security and other sustainability and governance objectives in the land sectors means that Article 6 could bring co-benefits that increase its attractiveness and the availability of finance, while also bringing risks that need to be monitored and mitigated against, such as uncertainties in measurements and the risk of non-permanence (Thamo and Pannell 2016⁵⁸⁷; Olsson et al. 2016⁵⁸⁸; Schwartz et al. 2017⁵⁸⁹). There has been progress in accounting for land-based emissions, mainly forestry and agriculture (medium evidence, low agreement), but various challenges remain (Macintosh 2012⁵⁹⁰; Pistorius et al. 2017⁵⁹¹; Krug 2018⁵⁹²).

Like the CDM and other existing carbon trading mechanisms, participation in Article 6.2 and 6.4 of the Paris Agreement requires certain institutional and data management capacities in the land sector to effectively benefit from the cooperation opportunities (Totin et al. 2018⁵⁹³). While the rules for the implementation of the new mechanisms are still under development, lessons from REDD+ (reducing emissions from deforestation and forest degradation) may be useful, which is perceived as more democratic and participative than the CDM (Maraseni and Cadman 2015⁵⁹⁴). Experience with REDD+ programmes emphasise the necessity to invest in ‘readiness’ programmes that assist countries to engage in strategic planning and build management and data collection systems to develop the capacity and infrastructure to participate in REDD+ (Minang et al. 2014⁵⁹⁵). The overwhelming majority of countries (93%) cite weak forest sector governance and institutions in their applications for REDD+ readiness funding (Kissinger et al. 2012⁵⁹⁶). Technology transfer for advanced remote sensing technologies that help to reduce uncertainty in monitoring forests helps to achieve REDD+ ‘readiness’ (Goetz et al. 2015⁵⁹⁷).

As well as new opportunities for finance and support, the Paris cooperation mechanisms and the associated roles for technology transfer bring new challenges, particularly in reporting, verifying and accounting in land-use sectors. Since developing countries must now achieve, measure and communicate emission reductions, they now have value for both developing and developed countries in achieving their NDCs, but reductions cannot be double-counted (i.e., towards multiple NDCs). All countries have to prepare and communicate NDCs, and many countries have included in their NDCs either economy-wide targets that include the land-use sectors, or specific targets for the land-use sectors. The Katowice climate package clarifies that all Parties have to submit ‘Biennial Transparency Reports’ from 2024 onwards, using common reporting formats, following most recent IPCC Guidelines (use of the 2013 Supplement on Wetlands is encouraged), identifying key categories of emissions, ensuring time-series consistency, and providing completeness and uncertainty assessments as well as quality control (UNFCCC 2018a⁵⁹⁸; Schneider and La Hoz Theuer 2019⁵⁹⁹). In total, the ambiguity in how countries incorporate land-use sectors into their NDC is estimated to lead to an uncertainty of more than 2 GtCO₂ in 2030 (Fyson and Jeffery 2018⁶⁰⁰). Uncertainty is lower if the analysis is limited to countries that have provided separate land-use sector targets in their NDCs (Benveniste et al. 2018⁶⁰¹).

Land-use zoning divides a territory (including local, sub-regional or national) into zones with different rules and regulations for land use (mining, agriculture, urban development, etc.), management practices and land-cover change (Metternicht 2018⁶⁴⁹). While the policy instrument is zoning ordinances, the process of determining these regulations is covered in integrated land-use planning (Section 7.6.2). Urban zoning can guide new growth in urban communities outside forecasted hazard areas, assist relocating existing dwellings to safer sites and manage post-event redevelopment in ways to reduce future vulnerability (Berke and Stevens 2016⁶⁵⁰). Holistic integration of climate mitigation and adaptation are interdependent and can be implemented by restoring urban forests, and improving parks (Brown 2010⁶⁵¹; Berke and Stevens 2016⁶⁵²). Zoning ordinances can contribute to SLM through protection of natural capital by preventing or limiting vegetation clearing, avoiding degradation of planning for rehabilitation of degraded land or contaminated sites, promoting conservation and enhancement of ecosystems and ecological corridors (Metternicht 2018⁶⁵³; Jepson and Haines 2014⁶⁵⁴). Zoning ordinances can also encourage higher density development, mixed use, local food production, encourage transportation alternatives (bike paths and transit-oriented development), preserve a sense of place, and increase housing diversity and affordability (Jepson and Haines 2014⁶⁵⁵). Conservation planning varies by context and may include one or several adaptation approaches, including protecting current patterns of biodiversity, large intact natural landscapes, and geophysical settings. Conservation planning may also maintain and restore ecological connectivity, identify and manage areas that provide future climate space for species expected to be displaced by climate change, and identify and protect climate refugia (Stevanovic et al. 2016⁶⁵⁶; Schmitz et al. 2015⁶⁵⁷).

Anguelovski et al. (2016)⁶⁵⁸ studied land-use interventions in eight cities in the global north and south, and concluded that historic trends of socio-economic vulnerability can be reinforced. They also found that vulnerability could be avoided with a consideration of the distribution of adaptation benefits and prioritising beneficial outcomes for disadvantaged and vulnerable groups when making future adaptation plans. Concentration of adaptation resources within wealthy business districts creating ecological enclaves exacerbated climate risks elsewhere and building of climate adaptive infrastructure such as sea walls or temporary flood barriers occurred at the expense of underserved neighbourhoods (Anguelovski et al. 2016a⁶⁵⁹).

There is limited evidence but high agreement that ecosystem-based adaptation (biodiversity, ecosystem services (ES), and Nature’s Contribution to People (see Chapter 6)) and incentives for ES – including payment for ecosystem services (PES) – play a critical part of an overall strategy to help people adapt to the adverse effects of climate change on land (UNEP 2009⁶⁶¹; Bonan 2008⁶⁶²; Millar et al. 2007⁶⁶³; Thompson et al. 2009⁶⁶⁴).

Ecosystem-based adaptation can promote socio-ecological resilience by enabling people to adapt to the impacts of climate change on land and reduce their vulnerability (Ojea 2015⁶⁶⁵). Ecosystem-based adaptation can promote nature conservation while alleviating poverty and even provide co-benefits by removing GHGs (Scarano 2017⁶⁶⁶) and protecting livelihoods (Munang et al. 2013⁶⁶⁷). For example, mangroves provide diverse ES such as carbon storage, fisheries, non-timber forest products, erosion protection, water purification, shore-line stabilisation, and also regulate storm surge and flooding damages, thus enhancing resilience and reducing climate risk from extreme events such as cyclones (Rahman et al. 2014⁶⁶⁸; Donato et al. 2011⁶⁶⁹; Das and Vincent 2009⁶⁷⁰; Ghosh et al. 2015⁶⁷¹; Ewel et al. 1998⁶⁷²).

There has been considerable increase in the last decade of PES, or programmes that exchange value for land management practices intended to ensure ES (Salzman et al. 2018⁶⁷³; Yang and Lu 2018⁶⁷⁴; Barbier 2011⁶⁷⁵). However, there is a deficiency in comprehensive and reliable data concerning the impact of PES on ecosystems, human well-being, their efficiency, and effectiveness (Pynegar et al. 2018⁶⁷⁶; Reed et al. 2014⁶⁷⁷; Salzman et al. 2018⁶⁷⁸; Barbier 2011⁶⁷⁹; Yang and Lu 2018⁶⁸⁰). While some studies assess ecological effectiveness and social equity, fewer assess economic efficiency (Yang and Lu 2018⁶⁸¹). Part of the challenge surrounds the fact that the majority of ES are not marketed, so determining how changes in ecosystems structures, functions and processes influence the quantity and quality of ES flows to people is challenging (Barbier 2011⁶⁸²). PES include agri-environmental targeted outcome-based payments, but challenges exist in relation to scientific uncertainty, pricing, timing of payments, increasing risk to land managers, World Trade Organization compliance, and barriers of land management and scale (Reed et al. 2014⁶⁸³).

PES is contested (Wang and Fu 2013⁶⁸⁴; Czembrowski and Kronenberg 2016⁶⁸⁵; Perry 2015⁶⁸⁶) for four reasons: (i) understanding and resolving trade-offs between conflicting groups of stakeholders (Wam et al. 2016⁶⁸⁷; Matthies et al. 2015⁶⁸⁸); (ii) knowledge and technology capacity (Menz et al. 2013⁶⁸⁹); (iii) challenges integrating PES with economic and other policy instruments (Ring and Schröter-Schlaack 2011⁶⁹⁰; Tallis et al. 2008⁶⁹¹; Elmqvist et al. 2003⁶⁹²; Albert et al. 2014⁶⁹³); and (iv) top-down climate change mitigation initiatives which are still largely carbon-centric, with limited opportunities for decentralised ecological restoration at local and regional scales (Vijge and Gupta 2014⁶⁹⁴).

During the past two decades, standards and certification have emerged as important sustainability and conservation instruments for agriculture, forestry, bioenergy, land-use management and bio-based products (Lambin et al. 2014⁷¹²; Englund and Berndes 2015⁷¹³; Milder et al. 2015⁷¹⁴; Giessen et al. 2016a⁷¹⁵; Endres et al. 2015⁷¹⁶; Byerlee et al. 2015⁷¹⁷; van Dam et al. 2010⁷¹⁸). Standards are normally voluntary, but can also become obligatory through legislation. A standard provides specifications or guidelines to ensure that materials, products, processes and services are fit for purpose, whereas certification is the procedure through which an accredited party confirms that a product, process or service is in conformity with certain standards. Standards and certification are normally carried out by separate organisations for legitimacy and accountability (Section7.6.6).The International Organization for Standardization is a key source for global environmental standards. Those with special relevance for land and climate include a recent standard on combating land degradation and desertification (ISO 2017⁷¹⁹) and an earlier standard on sustainable bioenergy and biomass use (ISO 2015⁷²⁰; Walter et al. 2018⁷²¹). Both aim to support the long-term transition to a climate-resilient bioeconomy; there is medium evidence on the sustainability implications of different bioeconomy pathways, but low agreement as to which pathways are socially and environmentally desirable (Priefer et al. 2017⁷²²; Johnson 2017⁷²³; Bennich et al. 2017a⁷²⁴).

Table 7.3 provides a summary of selected standards and certification schemes with a focus on land use and climate: the tickmark shows inclusion of different sustainability elements, with all recognising the inherent linkages between the biophysical and social aspects of land use. Some certification schemes and best practice guidelines are specific to a particular agriculture crop (e.g., soya, sugarcane) or a tree (e.g., oil palm) while others are general. International organisations promote sustainable land and biomass use through good practice guidelines, voluntary standards and jurisdictional approaches (Scarlat and Dallemand 2011⁷²⁵; Stattman et al. 2018a⁷²⁶). Other frameworks, such as the Global Bioenergy Partnership (GBEP) focus on monitoring land and biomass use through a set of indicators that are applied across partner countries, thereby also promoting technology/knowledge transfer (GBEP 2017⁷²⁷). The Economics of Land Degradation (ELD) Initiative provides common guidelines for economic assessments of land degradation (Nkonya et al. 2013⁷²⁸).

Whereas current standards and certification focus primarily on land, climate and biomass impacts where they occur, more recent analysis considers trade-related land-use change by tracing supply chain impacts from producer to consumer, leading to the notion of ‘imported deforestation’ that occurs from increasing demand and trade in unsustainable forest and agriculture products, which is estimated to account for 26% of all tropical deforestation (Pendrill et al. 2019⁷²⁹). Research and implementation efforts aim to improve supply chain transparency and promote commitments to ‘zero deforestation’ (Gardner et al. 2018a⁷³⁰; Garrett et al. 2019⁷³¹; Newton et al. 2018⁷³²; Godar and Gardner 2019⁷³³; Godar et al. 2015⁷³⁴, 2016). France has developed specific policies on imported deforestation that are expected to eventually include a ‘zero deforestation’ label (Government of France 2019).

Access to modern energy services is a key component of SDG 7, with an estimated 1.1 billion people lacking access to electricity, while nearly 3 billion people rely on traditional biomass (fuelwood, agriculture residues, animal dung, charcoal) for household energy needs (IEA 2017⁷⁵⁸). Lack of access to modern energy services is significant in the context of land-climate systems because heavy reliance on traditional biomass can contribute to land degradation, household air pollution and GHG emissions (see Cross-Chapter Box 12 in Chapter 7). A variety of policy instruments and programmes have been aimed at improving energy access and thereby reducing the heavy reliance on traditional biomass (Table 7.2); there is high evidence and high agreement that programmes and policies that reduce dependence on traditional biomass will have benefits for health and household productivity, as well as reducing land degradation (Section 4.5.4) and GHG emissions (Bailis et al. 2015⁷⁵⁹; Cutz et al. 2017a⁷⁶⁰; Masera et al. 2015⁷⁶¹; Goldemberg et al. 2018a⁷⁶²; Sola et al. 2016a⁷⁶³; Rao and Pachauri 2017⁷⁶⁴; Denton et al. 2014⁷⁶⁵). There can be trade-offs across different options, especially between health and climate benefits, since more efficient wood stoves might have only limited effect, whereas gaseous and liquid fuels (e.g., biogas, LPG, bioethanol) will have highly positive health benefits and climate benefits that vary depending on specific circumstances of the substitution (Cameron et al. 2016⁷⁶⁶; Goldemberg et al. 2018b⁷⁶⁷). Unlike traditional biomass, modern bioenergy offers high-quality energy services, although, for household cookstoves, even the cleanest options using wood may not perform as well in terms of health and/or climate benefits (Fuso Nerini et al. 2017⁷⁶⁸; Goldemberg et al. 2018b⁷⁶⁹).

There is a startling array of diverse and fragmented climate finance sources: more than 50 international public funds, 60 carbon markets, 6000 private equity funds, 99 multilateral and bilateral climate funds (Samuwai and Hills 2018⁷⁹⁹). Most public finance for developing countries flows through bilateral and multilateral institutions such as the World Bank, the International Monetary Fund, International Finance Corporation, regional development banks, as well as specialised multilateral institutions such as the Global Environmental Fund, and the EU Solidarity Fund. Some governments have established state investment banks (SIBs) to close the financing gap, including the UK (Green Investment Bank), Australia (Clean Energy Finance Corporation) and in Germany (Kreditanstalt für Wiederaufbau) the Development Bank has been involved in supporting low-carbon finance (Geddes et al. 2018⁸⁰⁰). The Green Climate Fund (GCF) now offers additional finance, but is still a new institution with policy gaps, a lengthy and cumbersome process related to approval (Brechin and Espinoza 2017⁸⁰¹; Khan and Roberts 2013⁸⁰²; Mathy and Blanchard 2016⁸⁰³), and challenges with adequate and sustained funding (Schalatek and Nakhooda 2013⁸⁰⁴). Private adaptation finance exists, but is difficult to define, track, and coordinate (Nakhooda et al. 2016⁸⁰⁵).

Comprehensive risk management (Section 7.4.3.1) designs a portfolio of instruments which are used across a continuum of preemptive, planning and assessment, and contingency measures in order to bolster resilience (Cummins and Weiss 2016⁸¹⁷) and address limitations of any one instrument (Surminski 2016⁸¹⁸; Surminski et al. 2016⁸¹⁹; Linnerooth-bayer et al. 2019⁸²⁰). Instruments designed and applied in isolation have shown short-term results, rather than sustained intended impacts (Vincent et al. 2018⁸²¹). Risk assessments limited to events and impacts on particular asset classes or sectors can misinform policy and drive misallocation of funding (Gallina et al. 2016⁸²²; Jongman et al. 2014⁸²³).

Comprehensive risk assessment combined with risk layering approaches that assign different instruments to different magnitude and frequency of events, have better potential to provide stability to societies facing disruption (Mechler et al. 2014⁸²⁴; Surminski et al. 2016⁸²⁵). Governments and citizens define limits of what they consider acceptable risks, risks for which market or other solutions can be developed and catastrophic risks that require additional public protection and intervention. Different financial tools may be used for these different categories of risk or phases of the risk cycle (preparedness, relief, recovery, reconstruction).

In order to protect lives and livelihoods early action is critical, including a coordinated plan for action agreed in advance, a fast, evidence-based decision-making process, and contingency financing to ensure that the plan can be implemented (Clarke and Dercon 2016a). Forecast-based finance mechanisms incorporate these principles, using climate or other indicators to trigger funding and action prior to a shock (Wilkinson 2018⁸²⁶). Forecast-based mechanisms can be linked with social protection systems by providing contingent scaled-up finance quickly to vulnerable populations following disasters, enhancing scalability, timeliness, predictability and adequacy of social protection benefits (Wilkinson 2018⁸²⁷; Costella et al. 2017b⁸²⁸; World Food Programme 2018⁸²⁹).

Measures in advance of risks set aside resources before negative impacts related to adverse weather, climatic stressors, and land changes occur. These tools are frequently applied in extreme event, rapid onset contexts. These measures are the main instruments for reducing fatalities and limiting damage from extreme climate and land change events (Surminski et al. 2016⁸³⁰). Finance tools in advance of risk include insurance (macro, meso, micro), green bonds, and forecast-based finance (Hunzai et al. 2018⁸³¹).

There is high confidence that insurance approaches that are designed to effectively reduce and communicate risks to the public and beneficiaries, designed to reduce risk and foster appropriate adaptive responses, and provide value in risk transfer, improve economic stability and social outcomes in both higher – and lower-income contexts (Kunreuther and Lyster 2016⁸³²; Outreville 2011b⁸³³; Surminski et al. 2016⁸³⁴; Kousky et al. 2018b⁸³⁵), bolster food security, help keep children in school, and help safeguard the ability of low-income households to pay for essentials like medicines (Shiferaw et al. 2014⁸³⁶; Hallegatte et al. 2017⁸³⁷).

Traditional financing mechanisms have not been sufficient and thereby leave a gap in facilitating a rapid transition to a low-carbon economy or building resilience (Geddes et al. 2018⁸⁷⁷). More recently there have been developments in more innovative mechanisms, including crowdfunding (Lam and Law 2016⁸⁷⁸), often supported by national governments (in the UK through regulatory and tax support) (Owen et al. 2018⁸⁷⁹). Crowdfunding has no financial intermediaries and thus low transaction costs, and the projects have a greater degree of independence than bank or institution funding (Miller et al. 2018⁸⁸⁰). Other examples of innovative mechanisms are community shares for local projects, such as renewable energy (Holstenkamp and Kahla 2016⁸⁸¹), or Corporate Power Purchase Agreements (PPAs) used by companies such as Google and Apple to purchase renewable energy directly or virtually from developers (Miller et al. 2018⁸⁸²). Investing companies benefit from avoiding unpredictable price fluctuations as well as increasing their environmental credentials. A second example is auctioned price floors, or subsidies that offer a guaranteed price for future emission reductions, currently being trialled in developing countries, by the World Bank Group, known as the Pilot Auction Facility for Methane and Climate Change Mitigation (PAF) (Bodnar et al. 2018⁸⁸³). Price floors can maximise the climate impact per public dollar while incentivising private investment in low-carbon technologies, and ideally would be implemented in conjunction with complementary policies such as carbon pricing.

In order for climate finance to be as effective and efficient as possible, cooperation between private, public and third sectors (e.g., non-governmental organisations (NGOs), cooperatives, and community groups) is more likely to create an enabling environment for innovation (Owen et al. 2018⁸⁸⁴). While innovative private sector approaches are making significant progress, the existence of a stable policy environment that provides certainty and incentives for long-term private investment is critical.

There are human, social, economic, and institutional barriers to adaptation to land-climate challenges as described in Table 7.4 (medium evidence, high agreement). Considerable literature exists around changing behaviours through response options targeting social and cultural barriers (Rosin 2013⁹³²; Eakin 2016⁹³³; Marshall et al. 2012⁹³⁴) (Chapter 6).

Since the publication of the IPCC’s Fifth Assessment Report (AR5) (IPCC 2014), research is emerging, examining the role of governance, institutions and (in particular) policy instruments, in creating or overcoming barriers to adaptation to land and climate change in the land-use sector (Foudi and Erdlenbruch 2012⁹³⁵; Linnerooth-Bayer and Hochrainer-Stigler 2015⁹³⁶). Evidence shows that understanding the local context and targeted approaches are generally most successful (Rauken et al. 2014⁹³⁷). Understanding the nature of constraints to adaptation is critical in determining how barriers may be overcome. Formal institutions (rules, laws, policies) and informal institutions (social and cultural norms and shared understandings) can be barriers and enablers of climate adaptation (Jantarasami et al. 2010⁹³⁸).

Governments play a key role in intervening and confronting existing barriers by changing legislation, adopting policy instruments, providing additional resources, and building institutions and knowledge exchange (Ford and Pearce 2010⁹³⁹; Measham et al. 2011⁹⁴⁰; Mozumder et al. 2011⁹⁴¹; Storbjörk 2010⁹⁴²). Understanding institutional barriers is important in addressing barriers (high confidence). Institutional barriers may exist due to the path-dependent nature of institutions governing natural resources and public good, bureaucratic structures that undermine horizontal and vertical integration (Section 7.6.2), and lack of policy coherence (Section 7.4.8).

Barriers to land-based mitigation relate to full understanding of the permanence of carbon sequestration in soils or terrestrial biomass, the additionality of this storage, its impact on production and production shifts to other regions, measurement and monitoring systems and costs (Smith et al. 2007⁹⁴³). Agricultural producers are more willing to expand mitigation measures already employed (including efficient and effective management of fertiliser, including manure and slurry) and less favourable to those not employed, such as using dietary additives, adopting genetically improved animals, or covering slurry tanks and lagoons (Feliciano et al. 2014⁹⁴⁴). Barriers identified in land- based mitigation include physical environmental constraints such as lack of information, education, and suitability for size and location of farm. For instance, precision agriculture is not viewed as efficient in small-scale farming (Feliciano et al. 2014⁹⁴⁵).

Property rights may be a barrier when there is no clear single- party land ownership to implement and manage changes (Smith et al. 2007⁹⁴⁶). In forestry, tenure arrangements may not distribute obligations and incentives for carbon sequestration effectively between public management agencies and private agents with forest licences. Including carbon in tenure and expanding the duration of tenure may provide stronger incentive for tenure holders to manage carbon as well as timber values (Williamson and Nelson 2017⁹⁴⁷). Effective policy will require answers as to the current status of agriculture in regard to GHG emissions, the degree that emissions are to change, the best pathway to achieve the change, and an ability to know when the target level of change is achieved (Smith et al. 2007⁹⁴⁸). Forest governance may not have the structure to advance mitigation and adaptation. Currently top-down traditional modes do not have the flexibility or responsiveness to deal with the complex, dynamic, spatially diverse, and uncertain features of climate change (Timberlake and Schultz 2017⁹⁴⁹; Williamson and Nelson 2017⁹⁵⁰).

In respect of forest mitigation, two main institutional barriers have been found to predominate. First, forest management institutions do not consider climate change to the degree necessary for enabling effective climate response, and do not link adaptation and mitigation. Second, institutional barriers exist if institutions are not forward looking, do not enable collaborative adaptive management, do not promote flexible approaches that are reversible as new information becomes available, do not promote learning and allow for diversity of approaches that can be tailored to different local circumstances (Williamson and Nelson 2017⁹⁵¹).

There is medium evidence and high agreement that one of the greatest challenges for land-based adaptation and SLM is posed by inequalities that influence vulnerability and coping and adaptive capacity – including age, gender, wealth, knowledge, access to resources and power (Kunreuther et al. 2014⁹⁵³; IPCC 2012⁹⁵⁴; Olsson et al. 2014⁹⁵⁵). Gender is the dimension of inequality that has been the focus of most research, while research demonstrating differential impacts, vulnerability and adaptive capacity based on age, ethnicity and indigeneity is less well developed (Olsson et al. 2015a⁹⁵⁶). Cross-Chapter Box 11 in Chapter 7 sets out both the contribution of gender relations to differential vulnerability and available policy instruments for greater gender inclusivity.

One response to the vulnerability of poor people and other categories differentially affected is effective and reliable social safety nets (Jones and Hiller 2017⁹⁵⁷). Social protection coverage is low across the world and informal support systems continue to be the key means of protection for a majority of the rural poor and vulnerable (Stavropoulou et al. 2017⁹⁵⁸) (Section 7.4.2). However, there is a gap in knowledge in understanding both positive and negative synergies between formal and informal systems of social protection and how local support institutions might be used to implement more formal forms of social protection (Stavropoulou et al. 2017⁹⁵⁹).

Inequalities of wealth and power can allow processes of corruption and elite capture (where public resources are used for the benefit of a few individuals in detriment to the larger populations) which can affect both adaptation and mitigation actions, at levels from the local to the global that, in turn, risk creating inequitable or unjust outcomes (Sovacool 2018⁹⁶⁰) (limited evidence, medium agreement). This includes risks of corruption in REDD+ processes (Sheng et al. 2016⁹⁶¹; Williams and Dupuy 2018⁹⁶²) and of corruption or elite capture in broader forest governance (Sundström 2016⁹⁶³; Persha and Andersson 2014⁹⁶⁴), as well as elite capture of benefits from planned adaptation at a local level (Sovacool 2018⁹⁶⁵).

Peer-reviewed empirical studies that focus on corruption in climate finance and interventions, particularly at a local level, are rare, due in part to the obvious difficulties of researching illegal and clandestine activity (Fadairo et al. 2017⁹⁶⁶). At the country level, historical levels of corruption are shown to affect current climate polices and global cooperation (Fredriksson and Neumayer 2016⁹⁶⁷). Brown (2010)⁹⁶⁸ sees three likely inlets of corruption into REDD+: in the setting of forest baselines, the reconciliation of project and natural credits, and the implementation of control of illegal logging. The transnational and north-south dimensions of corruption are highlighted by debates on which US legislative instruments (e.g., the Lacey Act, the Foreign Corrupt Practices Act) could be used to prosecute the northern corporations that are involved in illegal logging (Gordon 2016⁹⁶⁹; Waite 2011⁹⁷⁰).

Fadairo et al. (2017)⁹⁷¹ carried out a structured survey of perceptions of households in forest-edge communities served by REDD+, as well as those of local officials, in south eastern Nigeria. They report high rates of agreement that allocation of carbon rights is opaque and uncertain, distribution of benefits is untimely, uncertain and unpredictable, and the REDD+ decision-making process is vulnerable to political interference that benefits powerful individuals. Only 35% of respondents had an overall perception of transparency in REDD+ process as ‘good’. Of eight institutional processes or facilities previously identified by the government of Nigeria and international agencies as indicators of commitment to transparent and equitable governance, only three were evident in the local REDD+ office as ‘very functional’ or ‘fairly functional’.

At the local level, the risks of corruption and elite capture of the benefits of climate action are high in decentralised regimes (Persha and Andersson 2014⁹⁷²). Rahman (2018) discusses elicitation of bribes (by local-level government staff) and extortion (by criminals) to allow poor rural people to gather forest products. The results are a general undermining of households’ adaptive capacity and perverse incentives to over-exploit forests once bribes have been paid, leading to over-extraction and biodiversity loss. Where there are pre-existing inequalities and conflict, participation processes need careful management and firm external agency to achieve genuine transformation and avoid elite capture (Rigon 2014⁹⁷³). An illustration of the range of types of elite capture is given by Sovacool (2018)⁹⁷⁴ for adaptation initiatives including coastal afforestation, combining document review and key informant interviews in Bangladesh, with an analytical approach from political ecology. Four processes are discussed: enclosure, including land grabbing and preventing the poor establishing new land rights; exclusion of the poor from decision-making over adaptation; encroachment on the resources of the poor by new adaptation infrastructure; and entrenchment of community disempowerment through patronage. The article notes that observing these processes does not imply they are always present, nor that adaptation efforts should be abandoned.

Policy instruments that strengthen agricultural producer assets or capital reduce vulnerability and overcome barriers to adaptation (Hurlbert 2018b, 2015b⁹⁷⁵). Additional factors like formal education and knowledge of traditional farming systems, secure tenure rights, access to electricity and social institutions in rice-farming areas of Bangladesh have played a positive role in reducing adaptation barriers (Alam 2015⁹⁷⁶). A review of more than 168 publications over 15 years about adaptation of water resources for irrigation in Europe found the highest potential for action is in improving adaptive capacity and responding to changes in water demands, in conjunction with alterations in current water policy, farm extension training, and viable financial instruments (Iglesias and Garrote 2015⁹⁷⁷). Research on the Great Barrier Reef, the Olifants River in Southern Africa, and fisheries in Europe, North America, and the Antarctic Ocean, suggests that the leading factor in harnessing the adaptive capacity of ecosystems is to reduce human stressors by enabling actors to collaborate across diverse interests, institutional settings, and sectors (Biggs et al. 2017⁹⁷⁸; Schultz et al. 2015⁹⁷⁹; Johnson and Becker 2015⁹⁸⁰). Fostering equity and participation are correlated with the efficacy of local adaptation to secure food and livelihood security (Laube et al. 2012⁹⁸¹). In this chapter, we examine the literature surrounding appropriate policy instruments, decision-making, and governance practices to overcome limits and barriers to adaptation.

Incremental adaptation consists of actions where the central aim is to maintain the essence and integrity of a system or process at a given site, whereas transformational adaptation changes the fundamental attributes of a system in response to climate and its effects; the former is characterised as doing different things and the latter, doing things differently (Noble et al. 2014). Transformational adaptation is necessary in situations where there are hard limits to adaptation or it is desirable to address deficiencies in sustainability, adaptation, inclusive development and social equity (Kates et al. 2012⁹⁸²; Mapfumo et al. 2016⁹⁸³). In other situations, incremental changes may be sufficient (Hadarits et al. 2017⁹⁸⁴).

Formal decision-making processes can occur at all levels, including the global, regional, national and sub-national levels (Section 7.4.1). Formal decision-making support tools can be used, for example, by farmers, to answer ‘what-if’ questions as to how to respond to the effects of changing climate on soils, rainfall and other conditions (Wenkel et al. 2013¹⁰⁰⁰).

Optimal formal decision-making is based on realistic behaviour of actors, important in land–climate systems, assessed through participatory approaches, stakeholder consultations and by incorporating results from empirical analyses. Mathematical simulations and games (Lamarque et al. 2013¹⁰⁰¹), behavioural models in land-based sectors (Brown et al. 2017¹⁰⁰²), agent-based models and micro- simulations are examples useful to decision-makers (Bishop et al. 2013¹⁰⁰³). These decision-making tools are expanded on in Section 7.5.2.

There are different ways to incorporate local knowledge, informal institutions and other contextual characteristics that capture non- deterministic elements, as well as social and cultural beliefs and systems more generally, into formal decision-making (medium evidence, medium agreement) (Section 7.6.4). Classic scientific methodologies now include participatory and interdisciplinary methods and approaches (Jones et al. 2014¹⁰⁰⁴). Consequently, this broader range of approaches may capture informal and indigenous knowledge, improving the participation of indigenous peoples in decision-making processes, and thereby promote their rights to self-determination (Malogdos and Yujuico 2015b¹⁰⁰⁵) (Cross-Chapter Box 13 in Chapter 7).

Informal institutions have contributed to sustainable resources management (common pool resources) through creating a suitable environment for decision-making. The role of informal institutions indecision-making can be particularly relevant for land-use decisions and practices in rural areas in the global south and north (Huisheng 2015¹⁰⁰⁶). Understanding informal institutions is crucial for adapting to climate change, advancing technological adaptation measures, achieving comprehensive disaster management and advancing collective decision-making (Karim and Thiel 2017¹⁰⁰⁷). Informal institutions have been found to be a crucial entry point in dealing with vulnerability of communities and exclusionary tendencies impacting on marginalised and vulnerable people (Mubaya and Mafongoya 2017¹⁰⁰⁸).

Many studies underline the role of local/informal traditional institutions in the management of natural resources in different parts of the world (Yami et al. 2009¹⁰⁰⁹; Zoogah et al. 2015¹⁰¹⁰; Bratton 2007¹⁰¹¹; Mowo et al. 2013¹⁰¹²; Grzymala-Busse 2010¹⁰¹³). Traditional systems include: traditional silvopastoral management (Iran), management of rangeland resources (South Africa), natural resource management (Ethiopia, Tanzania, Bangladesh) communal grazing land management (Ethiopia) and management of conflict over natural resources (Siddig et al. 2007¹⁰¹⁴; Yami et al. 2011¹⁰¹⁵; Valipour et al. 2014¹⁰¹⁶; Bennett 2013¹⁰¹⁷; Mowo et al. 2013¹⁰¹⁸).

Formal–informal institutional interaction could take different shapes such as: complementary, accommodating, competing, and substitutive. There are many examples when formal institutions might obstruct, change, and hinder informal institutions (Rahman et al. 2014¹⁰¹⁹; Helmke and Levitsky 2004¹⁰²⁰; Bennett 2013¹⁰²¹; Osei-Tutu et al. 2014¹⁰²²). Similarly, informal institutions can replace, undermine, and reinforce formal institutions (Grzymala-Busse 2010). In the absence of formal institutions, informal institutions gain importance, requiring focus in relation to natural resources management and rights protection (Estrin and Prevezer 2011¹⁰²³; Helmke and Levitsky 2004¹⁰²⁴; Kangalawe et al. 2014¹⁰²⁵; Sauerwald and Peng 2013¹⁰²⁶; Zoogah et al. 2015¹⁰²⁷).

Community forestry comprises 22% of forests in tropical countries in contrast to large-scale industrial forestry (Hajjar et al. 2013¹⁰²⁸) and is managed with informal institutions, ensuring a sustainable flow of forest products and income, utilising traditional ecological knowledge to determine access to resources (Singh et al. 2018¹⁰²⁹). Policies that create an open platform for local debates and allow actors their own active formulation of rules strengthen informal institutions. Case studies in Zambia, Mali, Indonesia and Bolivia confirm that enabling factors for advancing the local ownership of resources and crafting durability of informal rules require recognition in laws, regulations and policies of the state (Haller et al. 2016¹⁰³⁰).

Structured decision-making occurs when there is scientific knowledge about cause and effect, little uncertainty, and agreement exists on values and norms relating to an issue (Hurlbert and Gupta 2016¹⁰³¹). This decision space is situated within the ‘known’ space where cause and effect is understood and predictable (although uncertainty is not quite zero) (French 2015¹⁰³²). Figure 7.5 displays the structured problem area in the bottom left-hand corner corresponding with the ‘known’ decision-making space. Decision-making surrounding quantified risk assessment and risk management (Section 7.4.3.1) occurs within this decision-making space. Examples in the land and climate area include cost-benefit analysis surrounding implementation of irrigation projects (Batie 2008¹⁰³³) or adopting soil erosion practices by agricultural producers based on anticipated profit (Hurlbert 2018b¹⁰³⁴). Comprehensive risk management also occupies this decision space (Papathoma-Köhle et al. 2016¹⁰³⁵), encompassing risk assessment, reduction, transfer, retention, emergency preparedness and response, and disaster recovery by combining quantified proactive and reactive approaches (Fra.Paleo 2015¹⁰³⁶) (Section 7.4.3).

Decisions can be made despite uncertainty (medium confidence), and a wide range of possible approaches are emerging to support decision-making under uncertainty (Jones et al. 2014¹⁰⁴²), applied both to adaptation and mitigation decisions.

Traditional approaches for economic appraisal, including cost- benefit analysis and cost-effectiveness analysis referred to in Section 7.5.2.1 do not handle or address uncertainty well (Hallegatte 2009¹⁰⁴³; Farber 2015¹⁰⁴⁴) and favour decisions with short-term benefits (see Cross-Chapter Box 10 in this chapter). Alternative economic decision-making approaches aim to better incorporate uncertainty while delivering adaptation goals, by selecting projects that meet their purpose across a variety of plausible futures (Hallegatte et al. 2012¹⁰⁴⁵) – so-called ‘robust’ decision-making approaches. These are designed to be less sensitive to uncertainty about the future (Lempert and Schlesinger 2000¹⁰⁴⁶).

Much of the research for adaptation to climate change has focused around three main economic approaches: real options analysis, portfolio analysis, and robust decision-making. Real options analysis develops flexible strategies that can be adjusted when additional climate information becomes available. It is most appropriate for large irreversible investment decisions. Applications to climate adaptation are growing quickly, with most studies addressing flood risk and sea-level rise (Gersonius et al. 2013¹⁰⁴⁷; Woodward et al. 2014¹⁰⁴⁸; Dan 2016¹⁰⁴⁹), but studies in land-use decisions are also emerging, including identifying the optimal time to switch land use in a changing climate (Sanderson et al. 2016¹⁰⁵⁰) and water storage (Sturm et al. 2017¹⁰⁵¹; Kim et al. 2017¹⁰⁵²). Portfolio analysis aims to reduce risk by diversification, by planting multiple species rather than only one, for example, in forestry (Knoke et al. 2017¹⁰⁵³) or crops (Ben-Ari and Makowski 2016¹⁰⁵⁴), or in multiple locations. There may be a trade- off between robustness to variability and optimality (Yousefpour and Hanewinkel 2016¹⁰⁵⁵; Ben-Ari and Makowski 2016¹⁰⁵⁶); but this type of analysis can help identify and quantify trade-offs. Robust decision-making identifies how different strategies perform under many climate outcomes, also potentially trading off optimality for resilience (Lempert 2013¹⁰⁵⁷).

Multi-criteria decision-making continues to be an important tool in the land-use sector, with the capacity to simultaneously consider multiple goals across different domains (e.g., economic, environmental, social) (Bausch et al. 2014¹⁰⁵⁸; Alrø et al. 2016¹⁰⁵⁹), and so is useful as a mitigation as well as an adaptation tool. Lifecycle assessment can also be used to evaluate emissions across a system – for example, in livestock production (McClelland et al. 2018¹⁰⁶⁰) – and to identify areas to prioritise for reductions. Bottom-up marginal abatement cost curves calculate the most cost effective cumulative potential for mitigation across different options (Eory et al. 2018¹⁰⁶¹).

In the climate adaptation literature, these tools may be used in adaptive management (Section 7.5.4), using a monitoring, research, evaluation and learning process (cycle) to improve future management strategies (Tompkins and Adger 2004¹⁰⁶²). More recently these techniques have been advanced with iterative risk management (IPCC 2014a¹⁰⁶³) (Sections 7.4.1 and 7.4.7), adaptation pathways (Downing 2012¹⁰⁶⁴), and dynamic adaptation pathways (Haasnoot et al. 2013¹⁰⁶⁵) (Section 7.6.3). Decision-making tools can be selected and adapted to fit the specific land and climate problem and decision- making space. For instance, dynamic adaptation pathways processes (Haasnoot et al. 2013¹⁰⁶⁶; Wise et al. 2014¹⁰⁶⁷) identify and sequence potential actions based on alternative potential futures and are situated within the complex, unstructured space (see Figure 7.5). Decisions are made based on trigger points, linked to indicators and scenarios, or changing performance over time (Kwakkel et al. 2016¹⁰⁶⁸). A key characteristic of these pathways is that, rather than making irreversible decisions now, decisions evolve over time, accounting for learning (Section 7.6.4), knowledge, and values. In New Zealand, combining dynamic adaptive pathways and a form of real options analysis with multiple-criteria decision analysis has enabled risk that changes over time to be included in the assessment of adaptation options through a participatory learning process (Lawrence et al. 2019¹⁰⁶⁹).

The Cross-Chapter Box 10 on Economic dimensions of climate change and land deals with the costs and timing of action. In terms of policies, not only is timing important, but the type of intervention itself can influence returns (high evidence, high agreement). Policy packages that make people more resilient – expanding financial inclusion, disaster risk and health insurance, social protection and adaptive safety nets, contingent finance and reserve funds, and universal access to early warning systems (Sections 7.4.1 and 7.6.3) – could save 100 billion USD a year, if implemented globally (Hallegatte et al. 2017¹⁰⁹⁵). In Ethiopia, Kenya and Somalia, every 1 USD spent on safety-net/resilience programming results in net benefits of between 2.3 and 3.3 USD (Venton 2018¹⁰⁹⁶). Investing in resilience-building activities, which increase household income by 365 to 450 USD per year in these countries, is more cost effective than providing ongoing humanitarian assistance.

There is a need to further examine returns on investment for land- based adaptation measures, both in the short and long term. Other outstanding questions include identifying specific triggers for early response. Food insecurity, for example, can occur due to a mixture of market and environmental factors (changes in food prices, animal or crop prices, rainfall patterns) (Venton 2018¹⁰⁹⁷). The efficacy of different triggers, intervention times and modes of funding are currently being evaluated (see, for example, forecast-based finance study; Alverson and Zommers 2018¹⁰⁹⁸). To reduce losses and maximise returns on investment, this information can be used to develop: 1) coordinated, agreed plans for action; 2) a clear, evidence-based decision-making process, and; 3) financing models to ensure that the plans for early action can be implemented (Clarke and Dercon 2016a¹⁰⁹⁹).

Unplanned or unintentional trade-offs and synergies between policy driven response options related to ecosystem services (ES) can happen over space (e.g., upstream-downstream, integrated watershed management, Section 3.7.5.2) or intensify over time (reduced water in future dry-season due to growing tree plantations, Section 6.4.1). Trade-offs can occur between two or more ES (land for climate mitigation vs food; Sections 6.2, 6.3, 6.4, Cross-Chapter Box 8 in Chapter 6; Cross-Chapter Box 9 in Chapters 6 and 7), and between scales, such as forest biomass-based livelihoods versus global ES carbon storage (Chhatre and Agrawal 2009¹¹⁷¹) (medium evidence, medium agreement). Trade-offs can be reversible or irreversible (Rodríguez et al. 2006¹¹⁷²; Elmqvist et al. 2013¹¹⁷³) (for example, a soil carbon sink is reversible) (Section 6.4.1.1).

Although there is robust evidence and high agreement that ES are important for human well-being, the relationship between poverty alleviation and ES can be surprisingly complex, understudied and dependent on the political economic context; current evidence is largely about provisioning services and often ignores multiple dimensions of poverty (Suich et al. 2015¹¹⁷⁴; Vira et al. 2012¹¹⁷⁵). Spatially explicit mapping and quantification of stakeholder choices in relation to distribution of various ES can help enhance synergies and reduce trade-offs (Turkelboom et al. 2018¹¹⁷⁶; Locatelli et al. 2014¹¹⁷⁷) (Section 7.5.5).

The Sustainable Development Goals (SDGs) are an international persuasive policy instrument that apply to all countries, and measure sustainable and socially just development of human societies at all scales of governance (Griggs et al. 2013¹¹⁷⁸). The UN SDGs rest on the premise that the goals are mutually reinforcing and there are inherent linkages, synergies and trade-offs (to a greater or lesser extent) between and within the sub-goals (Fuso Nerini et al. 2018¹¹⁷⁹; Nilsson et al. 2016b¹¹⁸⁰; Le Blanc 2015¹¹⁸¹). There is high confidence that opportunities, trade-offs and co-benefits are context – and region-specific and depend on a variety of political, national and socio-economic factors (Nilsson et al. 2016b¹¹⁸²) depending on perceived importance by decision-makers and policymakers (Figure 7.7 and Table 7.6). Aggregation of targets and indicators at the national level can mask severe biophysical and socio-economic trade-offs at local and regional scales (Wada et al. 2016¹¹⁸³).

There is medium evidence and high agreement that SDGs must not be pursued independently, but in a manner that recognises trade-offs and synergies with each other, consistent with a goal of ‘policy coherence’. Policy coherence also refers to spatial trade-offs and geopolitical implications within and between regions and countries implementing SDGs. For instance, supply-side food security initiatives of land-based agriculture are impacting on marine fisheries globally through creation of dead-zones due to agricultural run-off (Diaz and Rosenberg 2008¹¹⁸⁴).

SDGs 6 (clean water and sanitation), 7 (affordable and clean energy) and 9 (industry, innovation and infrastructure) are important SDGs related to mitigation with adaptation co-benefits, but they have local trade-offs with biodiversity and competing uses of land and rivers (see Case study: Green energy: Biodiversity conservation vs global environment targets) (medium evidence, high agreement) (Bogardi et al. 2012¹¹⁸⁵; Nilsson and Berggren 2000¹¹⁸⁶; Hoeinghaus et al. 2009¹¹⁸⁷; Winemiller et al. 2016¹¹⁸⁸). This has occurred despite emerging knowledge about the role that rivers and riverine ecosystems play in human development and in generating global, regional and local ES (Nilsson and Berggren 2000¹¹⁸⁹; Hoeinghaus et al. 2009¹¹⁹⁰). The transformation of river ecosystems for irrigation, hydropower and water requirements of societies worldwide is the biggest threat to freshwater and estuarine biodiversity and ecosystems services (Nilsson and Berggren 2000¹¹⁹¹; Vörösmarty et al. 2010¹¹⁹²). These projects address important energy and water-related demands, but their economic benefits are often overestimated in relation to trade-offs with respect to food (river capture fisheries), biodiversity and downstream ES (Winemiller et al. 2016¹¹⁹³). Some trade-offs and synergies related to SDG7 impact on aspirations of greater welfare and well-being, as well as physical and social infrastructure for sustainable development (Fuso Nerini et al. 2018¹¹⁹⁴) (Section 7.5.6.1, where trade-offs exist between climate mitigation and food).

Retaining existing forests, restoring degraded forest and afforestation are response options for climate change mitigation with adaptation benefits (Section 6.4.1). Policies at various levels of governance that foster ownership, autonomy, and provide incentives for forest cover can reduce trade-offs between carbon sinks in forests and local livelihoods (especially when the size of forest commons is sufficiently large) (Chhatre and Agrawal 2009¹²¹⁶; Locatelli et al. 2014¹²¹⁷) (see Table 7.6 this section; Case study: Forest conservation instruments: REDD+ in the Amazon and India, Section 7.4.6).

Trade-offs between some types of water use (e.g., irrigation for food security) and other ecosystem services (ES) are expected to intensify under climate change (Hanjra and Ejaz Qureshi 2010¹²⁴³). There is an urgency to develop approaches to understand and communicate this to policymakers and decision-makers (Zheng et al. 2016¹²⁴⁴). Reducing water use in agriculture (Mekonnen and Hoekstra 2016¹²⁴⁵) through policies on both the supply and demand side, such as a shift to less water-intensive crops (Richter et al. 2017¹²⁴⁶; Fishman et al. 2015¹²⁴⁷), and a shift in diets (Springmann et al. 2016¹²⁴⁸) has the potential to reduce trade-offs between food security and freshwater aquatic ES (medium evidence, high agreement). There is strong evidence that improved efficiency in irrigation can actually increase overall water use in agriculture, and therefore its contribution to improved flows in rivers is questionable (Ward and Pulido-Velazquez 2008¹²⁴⁹).

Coherent policies that consider synergies and trade-offs can also reduce the likelihood of maladaptation, which is the opposite of sustainable adaptation (Magnan et al. 2016¹²⁵⁴). Sustainable adaptation ‘contributes to socially and environmentally sustainable development pathways including both social justice and environmental integrity’ (Eriksen et al. 2011¹²⁵⁵). In IPCC’s Fifth Assessment Report (AR5) there was medium evidence and high agreement that maladaptation is ‘a cause of increasing concern to adaptation planners, where intervention in one location or sector could increase the vulnerability of another location or sector, or increase the vulnerability of a group to future climate change’ (Noble et al. 2014¹²⁵⁶). AR5 recognised that maladaptation arises not only from inadvertent, badly planned adaptation actions, but also from deliberate decisions where wider considerations place greater emphasis on short-term outcomes ahead of longer-term threats, or that discount, or fail to consider, the full range of interactions arising from planned actions (Noble et al. 2014¹²⁵⁷).

Some maladaptations are only beginning to be recognised as we become aware of unintended consequences of decisions. An example prevalent across many countries is irrigation as an adaptation to water scarcity. During a drought from 2007–2009 in California, farmers adapted by using more groundwater, thereby depleting groundwater elevation by 15 metres. This volume of groundwater depletion is unsustainable environmentally and also emits GHG emissions during the pumping (Christian-Smith et al. 2015¹²⁵⁸). Despite the three years of drought, the agricultural sector performed financially well, due to the groundwater use and crop insurance payments. Drought compensation programmes through crop insurance policies may reduce the incentive to shift to lower water-use crops, thereby perpetuating the maladaptive situation. Another example of maladaptation that may appear as adaptation to drought is pumping out groundwater and storing in surface farm ponds, with consequences for water justice, inequity and sustainability (Kale 2017¹²⁵⁹). These examples highlight the potential for maladaptation from farmers’ adaptation decisions as well as the unintended consequences of policy choices; the examples illustrate the findings of Barnett and O’Neill (2010)¹²⁶⁰ that maladaptation can include: high opportunity costs (including economic, environmental, and social); reduced incentives to adapt (adaptation measures that reduce incentives to adapt by not addressing underlying causes); and path dependency or trajectories that are difficult to change.

In practice, maladaptation is a specific instance of policy incoherence, and it may be useful to develop a framework in designing policy to avoid this type of trade-off. This would specify the type, aim and target audience of an adaptation action, decision, project, plan, or policy designed initially for adaptation, but actually at high risk of inducing adverse effects, either on the system in which it was developed, or another connected system, or both. The assessment requires identifying system boundaries, including temporal and geographical scales at which the outcomes are assessed (Magnan 2014¹²⁶¹; Juhola et al. 2016¹²⁶²). National-level institutions that cover the spectrum of sectors affected, or enhanced collaboration between relevant institutions, is expected to increase the effectiveness of policy instruments, as are joint programmes and funds (Morita and Matsumoto 2018¹²⁶³).

As new knowledge about trade-offs and synergies amongst land- climate processes emerges regionally and globally, concerns over emerging risks and the need for planning policy responses grow. There is medium evidence and medium agreement that trade- offs currently do not figure into existing climate policies including NDCs and SDGs being vigorously pursued by some countries (Woolf et al. 2018¹²⁶⁴). For instance, the biogeophysical co-benefits of reduced deforestation and re/afforestation measures (Chapter 6) are usually not accounted for in current climate policies or in the NDCs, but there is increasing scientific evidence to include them as part of the policy design (Findell et al. 2017¹²⁶⁵; Hirsch et al. 2018¹²⁶⁶; Bright et al. 2017¹²⁶⁷).