Editing IPCC:AR6/SRCCL/Chapter-4 (section)

== 4.8 4.8 Addressing land degradation in the context of climate change ==

<div id="article-4-8-addressing-land-degradation-in-the-context-of-climate-change-block-1"></div>

Land degradation in the form of soil carbon loss is estimated to have been ongoing for at least 12,000 years, but increased exponentially in the last 200 years (Sanderman et al. 2017 <sup>[[#fn:r912|912]]</sup> ). Before the advent of modern sources of nutrients, it was imperative for farmers to maintain and improve soil fertility through the prevention of runoff and erosion, and management of nutrients through vegetation residues and manure. Many ancient farming systems were sustainable for hundreds and even thousands of years, such as raised-field agriculture in Mexico (Crews and Gliessman 1991 <sup>[[#fn:r913|913]]</sup> ), tropical forest gardens in Southeast Asia and Central America (Ross 2011 <sup>[[#fn:r914|914]]</sup> ; Torquebiau 1992 <sup>[[#fn:r915|915]]</sup> ; Turner and Sabloff 2012 <sup>[[#fn:r916|916]]</sup> ), terraced agriculture in East Africa, Central America, Southeast Asia and the Mediterranean basin (Turner and Sabloff 2012 <sup>[[#fn:r917|917]]</sup> ; Preti and Romano 2014 <sup>[[#fn:r918|918]]</sup> ; Widgren and Sutton 2004 <sup>[[#fn:r919|919]]</sup> ; Håkansson and Widgren 2007 <sup>[[#fn:r920|920]]</sup> ; Davies and Moore 2016 <sup>[[#fn:r921|921]]</sup> ; Davies 2015 <sup>[[#fn:r922|922]]</sup> ), and integrated rice–fish cultivation in East Asia (Frei and Becker 2005 <sup>[[#fn:r923|923]]</sup> ).

Such long-term sustainable farming systems evolved in very different times and geographical contexts, but they share many common features, such as: the combination of species and structural diversity in time and space (horizontally and vertically) in order to optimise the use of available land; recycling of nutrients through biodiversity of plants, animals and microbes; harnessing the full range of site-specific micro-environments (e.g., wet and dry soils); biological interdependencies which help suppression of pests; reliance on mainly local resources; reliance on local varieties of crops, and sometimes incorporation of wild plants and animals; the systems are often labour and knowledge intensive (Rudel et al. 2016 <sup>[[#fn:r924|924]]</sup> ; Beets 1990 <sup>[[#fn:r925|925]]</sup> ; Netting 1993 <sup>[[#fn:r926|926]]</sup> ; Altieri and Koohafkan 2008 <sup>[[#fn:r927|927]]</sup> ). Such farming systems have stood the test of time and can provide important knowledge for adapting farming systems to climate change (Koohafkann and Altieri 2011 <sup>[[#fn:r928|928]]</sup> ).

In modern agriculture, the importance of maintaining the biological productivity and ecological integrity of farmland has not been a necessity in the same way as in pre-modern agriculture because nutrients and water have been supplied externally. The extreme land degradation in the US Midwest during the Dust Bowl period in the 1930s became an important wake-up call for agriculture and agricultural research and development, from which we can still learn much in order to adapt to ongoing and future climate change (McLeman et al. 2014 <sup>[[#fn:r929|929]]</sup> ; Baveye et al. 2011 <sup>[[#fn:r930|930]]</sup> ; McLeman and Smit 2006 <sup>[[#fn:r931|931]]</sup> ).

SLM is a unifying framework for addressing land degradation and can be defined as the stewardship and use of land resources, including soils, water, animals and plants, to meet changing human needs, while simultaneously ensuring the long-term productive potential of these resources and the maintenance of their environmental functions. It is a comprehensive approach comprising technologies combined with social, economic and political enabling conditions (Nkonya et al. 2011 <sup>[[#fn:r932|932]]</sup> ). It is important to stress that farming systems are informed by both scientific and local/traditional knowledge. The power of SLM in small-scale diverse farming was demonstrated effectively in Nicaragua after the severe cyclone Mitch in 1998 (Holt-Giménez 2002 <sup>[[#fn:r933|933]]</sup> ). Pairwise analysis of 880 fields with and without implementation of SLM practices showed that the SLM fields systematically fared better than the fields without SLM in terms of more topsoil remaining, higher field moisture, more vegetation, less erosion and lower economic losses after the cyclone. Furthermore, the difference between fields with and without SLM increased with increasing levels of storm intensity, slope gradient, and age of SLM (Holt-Giménez 2002 <sup>[[#fn:r934|934]]</sup> ).

When addressing land degradation through SLM and other approaches, it is important to consider feedbacks that impact on climate change. Table 4.2 shows some of the most important land degradation issues, their potential solutions, and their impacts on climate change. This table provides a link between the comprehensive lists of land degradation processes (Table 4.1) and land management solutions.

<div id="article-4-8-addressing-land-degradation-in-the-context-of-climate-change-block-2"></div>

<span id="table-4.2"></span>
<!-- START IMG -->
<!-- TABLE IMG -->
<!-- IMG TITLE -->
'''Table 4.2'''

<span id="interaction-of-human-and-climate-drivers-can-exacerbate-desertification-and-land-degradation."></span>
<!-- IMG CAPTION -->
 '''Interaction of human and climate drivers can exacerbate desertification and land degradation.'''

Climate change exacerbates the rate and magnitude of several ongoing land degradation and desertification processes. Human drivers of land degradation and desertification include expanding agriculture, agricultural practices and forest management. In turn, land degradation and desertification are also drivers of climate change through GHG emissions, reduced rates of carbon uptake, and reduced capacity of ecosystems to act as carbon sinks into the future. Impacts on climate change are either warming (in red) or cooling (in blue).
<!-- IMG FILE -->
[[File:4a018c823d8fff085acb70cdec8a6a2c table-4.2-a.png]] [[File:b624f630b9ff0bbdc71c1306a72f626d table-4.2-b.png]]

<!-- END IMG -->
<span id="actions-on-the-ground-to-address-land-degradation"></span>
=== 4.8.1 4.8.1 Actions on the ground to address land degradation ===

<div id="section-4-8-1-actions-on-the-ground-to-address-land-degradation-block-1"></div>

Concrete actions on the ground to address land degradation are primarily focused on soil and water conservation. In the context of adaptation to climate change, actions relevant for addressing land degradation are sometimes framed as ecosystem-based adaptation (Scarano 2017 <sup>[[#fn:r935|935]]</sup> ) or Nature-Based Solutions (Nesshöver et al. 2017 <sup>[[#fn:r936|936]]</sup> ), and in an agricultural context, agroecology (see Glossary) provides an important frame. The site-specific biophysical and social conditions, including local and indigenous knowledge, are important for successful implementation of concrete actions.

Responses to land degradation generally take the form of agronomic measures (methods related to managing the vegetation cover), soil management (methods related to tillage, nutrient supply), and mechanical methods (methods resulting in durable changes to the landscape) (Morgan 2005a <sup>[[#fn:r937|937]]</sup> ). Measures may be combined to reinforce benefits to land quality, as well as improving carbon sequestration that supports climate change mitigation. Some measures offer adaptation options and other co-benefits, such as agroforestry, involving planting fruit trees that can support food security in the face of climate change impacts (Reed and Stringer 2016 <sup>[[#fn:r938|938]]</sup> ), or application of compost or biochar that enhances soil water-holding capacity, so increases resilience to drought.

There are important differences in terms of labour and capital requirements for different technologies, and also implications for land tenure arrangements. Agronomic measures and soil management require generally little extra capital input and comprise activities repeated annually, so have no particular implication for land tenure arrangements. Mechanical methods require substantial upfront investments in terms of capital and labour, resulting in long-lasting structural change, requiring more secure land tenure arrangements (Mekuriaw et al. 2018 <sup>[[#fn:r939|939]]</sup> ). Agroforestry is a particularly important strategy for SLM in the context of climate change because of the large potential to sequester carbon in plants and soil and enhance resilience of agricultural systems (Zomer et al. 2016 <sup>[[#fn:r940|940]]</sup> ).

Implementation of SLM practices has been shown to increase the productivity of land (Branca et al. 2013 <sup>[[#fn:r941|941]]</sup> ) and to provide good economic returns on investment in many different settings around the world (Mirzabaev et al. 2015 <sup>[[#fn:r942|942]]</sup> ). Giger et al. (2018) <sup>[[#fn:r943|943]]</sup> showed, in a meta study of 363 SLM projects over the period 1990 to 2012, that 73% of the projects were perceived to have a positive or at least neutral cost-benefit ratio in the short term, and 97% were perceived to have a positive or very positive cost-benefit ratio in the long term ( ''robust evidence, high agreement'' ). Despite the positive effects, uptake is far from universal. Local factors, both biophysical conditions (e.g., soils,

drainage, and topography) and socio-economic conditions (e.g., land tenure, economic status, and land fragmentation) play decisive roles in the interest in, capacity to undertake, and successful implementation of SLM practices (Teshome et al. 2016 <sup>[[#fn:r944|944]]</sup> ; Vogl et al. 2017 <sup>[[#fn:r945|945]]</sup> ; Tesfaye et al. 2016 <sup>[[#fn:r946|946]]</sup> ; Cerdà et al. 2018 <sup>[[#fn:r947|947]]</sup> ; Adimassu et al. 2016 <sup>[[#fn:r948|948]]</sup> ). From a landscape perspective, SLM can generate benefits, including adaptation to and mitigation of climate change, for entire watersheds, but challenges remain regarding coordinated and consistent implementation ( ''medium evidence, medium agreement'' ) (Kerr et al. 2016 <sup>[[#fn:r949|949]]</sup> ; Wang et al. 2016a <sup>[[#fn:r950|950]]</sup> ).

<div id="section-4-8-1-1-agronomic-and-soil-management-measures"></div>

<span id="agronomic-and-soil-management-measures"></span>
==== 4.8.1.1 4.8.1.1 Agronomic and soil management measures ====

<div id="section-4-8-1-1-agronomic-and-soil-management-measures-block-1"></div>

Rebuilding soil carbon is an important goal of SLM, particularly in the context of climate change (Rumpel et al. 2018 <sup>[[#fn:r951|951]]</sup> ). The two most important reasons why agricultural soils have lost 20–60% of the soil carbon they contained under natural ecosystem conditions are the frequent disturbance through tillage and harvesting, and the change from deep- rooted perennial plants to shallow-rooted annual plants (Crews and Rumsey 2017 <sup>[[#fn:r952|952]]</sup> ). Practices that build soil carbon are those that increase organic matter input to soil, or reduce decomposition of SOM.

Agronomic practices can alter the carbon balance significantly, by increasing organic inputs from litter and roots into the soil. Practices include retention of residues, use of locally adapted varieties, inter-cropping, crop rotations, and green manure crops that replace the bare field fallow during winter and are eventually ploughed before sowing the next main crop (Henry et al. 2018 <sup>[[#fn:r953|953]]</sup> ). Cover crops (green manure crops and catch crops that are grown between the main cropping seasons) can increase soil carbon stock by between 0.22 and 0.4 t C ha <sup>–1</sup> yr <sup>–1</sup> (Poeplau and Don 2015 <sup>[[#fn:r954|954]]</sup> ; Kaye and Quemada 2017 <sup>[[#fn:r955|955]]</sup> ).

Reduced tillage (or no-tillage) is an important strategy for reducing soil erosion and nutrient loss by wind and water (Van Pelt et al. 2017 <sup>[[#fn:r956|956]]</sup> ; Panagos et al. 2015 <sup>[[#fn:r957|957]]</sup> ; Borrelli et al. 2016 <sup>[[#fn:r958|958]]</sup> ). But the evidence that no-till agriculture also sequesters carbon is not compelling (VandenBygaart 2016 <sup>[[#fn:r959|959]]</sup> ). Soil sampling of only the upper 30 cm can give biased results, suggesting that soils under no-till practices have higher carbon content than soils under conventional tillage (Baker et al. 2007 <sup>[[#fn:r960|960]]</sup> ; Ogle et al. 2012 <sup>[[#fn:r961|961]]</sup> ; Fargione et al. 2018 <sup>[[#fn:r962|962]]</sup> ; VandenBygaart 2016 <sup>[[#fn:r963|963]]</sup> ).

Changing from annual to perennial crops can increase soil carbon content (Culman et al. 2013 <sup>[[#fn:r964|964]]</sup> ; Sainju et al. 2017 <sup>[[#fn:r965|965]]</sup> ). A perennial grain crop (intermediate wheatgrass) was, on average, over four years a net carbon sink of about 13.5 tCO <sub>2</sub> ha <sup>–1</sup> yr <sup>–1</sup> (de Oliveira et al. 2018 <sup>[[#fn:r966|966]]</sup> ). Sprunger et al. (2018) <sup>[[#fn:r967|967]]</sup> compared an annual winter wheat crop with a perennial grain crop (intermediate wheatgrass) and found that the perennial grain root biomass was 15 times larger than winter wheat, however, there was no significant difference in soil carbon pools after the four-year experiment. Exactly how much, and over what time period, carbon can be sequestered through changing from annual to perennial crops depends on the degree of soil carbon depletion and other local biophysical factors (Section 4.9.2).

Integrated soil fertility management is a sustainable approach to nutrient management that uses a combination of chemical and organic amendments (manure, compost, biosolids, biochar), rhizobial nitrogen fixation, and liming materials to address soil chemical constraints (Henry et al. 2018 <sup>[[#fn:r968|968]]</sup> ). In pasture systems, management of grazing pressure, fertilisation, diverse species including legumes and perennial grasses can reduce erosion and enhance soil carbon (Conant et al. 2017 <sup>[[#fn:r969|969]]</sup> ).

<div id="section-4-8-1-2-mechanical-soil-and-water-conservation"></div>

<span id="mechanical-soil-and-water-conservation"></span>
==== 4.8.1.2 Mechanical soil and water conservation ====

<div id="section-4-8-1-2-mechanical-soil-and-water-conservation-block-1"></div>

In hilly and mountainous terrain, terracing is an ancient but still practised soil conservation method worldwide (Preti and Romano 2014 <sup>[[#fn:r970|970]]</sup> ) in climatic zones from arid to humid tropics (Balbo 2017 <sup>[[#fn:r981|981]]</sup> ). By reducing the slope gradient of hillsides, terraces provide flat surfaces. Deep, loose soils that increase infiltration, reduce erosion and thus sediment transport. They also decrease the hydrological connectivity and thus reduce hillside runoff (Preti et al. 2018 <sup>[[#fn:r972|972]]</sup> ; Wei et al. 2016 <sup>[[#fn:r973|973]]</sup> ; Arnáez et al. 2015 <sup>[[#fn:r974|974]]</sup> ; Chen et al. 2017 <sup>[[#fn:r975|975]]</sup> ). In terms of climate change, terraces are a form of adaptation that helps in cases where rainfall is increasing or intensifying (by reducing slope gradient and the hydrological connectivity), and where rainfall is decreasing (by increasing infiltration and reducing runoff) ( ''robust evidence, high agreement'' ). There are several challenges, however, to continued maintenance and construction of new terraces, such as the high costs in terms of labour and/or capital (Arnáez et al. 2015 <sup>[[#fn:r976|976]]</sup> ) and disappearing local knowledge for maintaining and constructing new terraces (Chen et al. 2017 <sup>[[#fn:r977|977]]</sup> ). The propensity of farmers to invest in mechanical soil conservation methods varies with land tenure; farmers with secure tenure arrangements are more willing to invest in durable practices such as terraces (Lovo 2016 <sup>[[#fn:r978|978]]</sup> ; Sklenicka et al. 2015 <sup>[[#fn:r979|979]]</sup> ; Haregeweyn et al. 2015 <sup>[[#fn:r980|980]]</sup> ). Where the slope is less severe, erosion can be controlled by contour banks, and the keyline approach (Duncan 2016 <sup>[[#fn:r1652|1652]]</sup> ; Stevens et al. 2015 <sup>[[#fn:r982|982]]</sup> ) to soil and water conservation.

<div id="section-4-8-1-3-agroforestry"></div>

<span id="agroforestry"></span>
==== 4.8.1.3 Agroforestry ====

<div id="section-4-8-1-3-agroforestry-block-1"></div>

Agroforestry is defined as a collective name for land-use systems in which woody perennials (trees, shrubs, etc.) are grown in association with herbaceous plants (crops, pastures) and/or livestock in a spatial arrangement, a rotation, or both, and in which there are both ecological and economic interactions between the tree and non-tree components of the system (Young, 1995, p. 11 <sup>[[#fn:r983|983]]</sup> ). At least since the 1980s, agroforestry has been widely touted as an ideal land management practice in areas vulnerable to climate variations and subject to soil erosion. Agroforestry holds the promise of improving soil and climatic conditions, while generating income from wood energy, timber and non-timber products – sometimes presented as a synergy of adaptation and mitigation of climate change (Mbow et al. 2014 <sup>[[#fn:r984|984]]</sup> ).

There is strong scientific consensus that a combination of forestry with agricultural crops and/or livestock, agroforestry systems can provide additional ecosystem services when compared with monoculture crop systems (Waldron et al. 2017 <sup>[[#fn:r985|985]]</sup> ; Sonwa et al. 2011 <sup>[[#fn:r986|986]]</sup> , 2014 <sup>[[#fn:r987|987]]</sup> , 2017 <sup>[[#fn:r988|988]]</sup> ; Charles et al. 2013 <sup>[[#fn:r989|989]]</sup> ). Agroforestry can enable sustainable intensification by allowing continuous production on the same unit of land with higher productivity without the need to use shifting agriculture systems to maintain crop yields (Nath et al. 2016 <sup>[[#fn:r990|990]]</sup> ). This is especially relevant where there is a regional requirement to find a balance between the demand for increased agricultural production and the protection of adjacent natural ecosystems such as primary and secondary forests (Mbow et al. 2014 <sup>[[#fn:r991|991]]</sup> ). For example, the use of agroforestry for perennial crops such as coffee and cocoa is increasingly promoted as offering a route to sustainable farming, with important climate change adaptation and mitigation co-benefits (Sonwa et al. 2001 <sup>[[#fn:r992|992]]</sup> ; Kroeger et al. 2017 <sup>[[#fn:r993|993]]</sup> ). Reported co-benefits of agroforestry in cocoa production include increased carbon sequestration in soils and biomass, improved water and nutrient use efficiency and the creation of a favourable micro-climate for crop production (Sonwa et al. 2017 <sup>[[#fn:r994|994]]</sup> ; Chia et al. 2016 <sup>[[#fn:r995|995]]</sup> ). Importantly, the maintenance of soil fertility using agroforestry has the potential to reduce the practice of shifting agriculture (of cocoa) which results in deforestation (Gockowski and Sonwa 2011 <sup>[[#fn:r996|996]]</sup> ). However, positive interactions within these systems can be ecosystem and/or species specific, but co-benefits such as increased resilience to extreme climate events, or improved soil fertility are not always observed (Blaser et al. 2017 <sup>[[#fn:r997|997]]</sup> ; Abdulai et al. 2018 <sup>[[#fn:r998|998]]</sup> ). These contrasting outcomes indicate the importance of field-scale research programmes to inform agroforestry system design, species selection and management practices (Sonwa et al. 2014 <sup>[[#fn:r999|999]]</sup> ).

Despite the many proven benefits, adoption of agroforestry has been low and slow (Toth et al. 2017 <sup>[[#fn:r1000|1000]]</sup> ; Pattanayak et al. 2003 <sup>[[#fn:r1001|1001]]</sup> ; Jerneck and Olsson 2014 <sup>[[#fn:r1002|1002]]</sup> ). There are several reasons for the slow uptake, but the perception of risks and the time lag between adoption and realisation of benefits are often important (Pattanayak et al. 2003 <sup>[[#fn:r1003|1003]]</sup> ; Mercer 2004 <sup>[[#fn:r1004|1004]]</sup> ; Jerneck and Olsson 2013 <sup>[[#fn:r1005|1005]]</sup> ).

An important question for agroforestry is whether it supports poverty alleviation, or if it favours comparatively affluent households. Experiences from India suggest that the overall adoption is low, with a differential between rich and poor households. Brockington el al. (2016) <sup>[[#fn:r1006|1006]]</sup> , studied agroforestry adoption over many years in South India and found that, overall, only 18% of the households adopted agroforestry. However, among the relatively rich households who adopted agroforestry, 97% were still practising it after six to eight years, and some had expanded their operations. Similar results were obtained in Western Kenya, where food-secure households were much more willing to adopt agroforestry than food-insecure households (Jerneck and Olsson 2013 <sup>[[#fn:r1007|1007]]</sup> , 2014). Other experiences from Sub-Saharan Africa illustrate the difficulties (such as local institutional support) of having a continued engagement of communities in agroforestry (Noordin et al. 2001 <sup>[[#fn:r1008|1008]]</sup> ; Matata et al. 2013 <sup>[[#fn:r1009|1009]]</sup> ; Meijer et al. 2015 <sup>[[#fn:r1010|1010]]</sup> ).

<div id="section-4-8-1-4-crop-livestock-interaction-as-an-approach-to-managing-land-degradation"></div>

<span id="croplivestock-interaction-as-an-approach-to-managing-land-degradation"></span>
==== 4.8.1.4 Crop–livestock interaction as an approach to managing land degradation ====

<div id="section-4-8-1-4-crop-livestock-interaction-as-an-approach-to-managing-land-degradation-block-1"></div>

The integration of crop and livestock production into ‘mixed farming’ for smallholders in developing countries became an influential model, particularly for Africa, in the early 1990s (Pritchard et al. 1992 <sup>[[#fn:r1011|1011]]</sup> ; McIntire et al. 1992 <sup>[[#fn:r1012|1012]]</sup> ). Crop–livestock integration under this model was seen as founded on three pillars: improved use of manure for crop fertility management; expanded use of animal traction (draught animals); and promotion of cultivated fodder crops. For Asia, emphasis was placed on draught power for land preparation, manure for soil fertility enhancement, and fodder production as an entry point for cultivation of legumes (Devendra and Thomas 2002 <sup>[[#fn:r1013|1013]]</sup> ). Mixed farming was seen as an evolutionary process to expand food production in the face of population increase, promote improvements in income and welfare, and protect the environment. The process could be further facilitated and steered by research, agricultural advisory services and policy (Pritchard et al. 1992 <sup>[[#fn:r1014|1014]]</sup> ; McIntire et al. 1992 <sup>[[#fn:r1015|1015]]</sup> ; Devendra 2002 <sup>[[#fn:r1016|1016]]</sup> ).

Scoones and Wolmer (2002) <sup>[[#fn:r1017|1017]]</sup> place this model in historical context, including concern about population pressure on resources and the view that mobile pastoralism was environmentally damaging. The latter view had already been critiqued by developing understandings of pastoralism, mobility and communal tenure of grazing lands (e.g., Behnke 1994 <sup>[[#fn:r1018|1018]]</sup> ; Ellis 1994 <sup>[[#fn:r1019|1019]]</sup> ). They set out a much more differentiated picture of crop–livestock interactions, which can take place either within a single-farm household, or between crop and livestock producers, in which case they will be mediated by formal and informal institutions governing the allocation of land, labour and capital, with the interactions evolving through multiple place-specific pathways (Ramisch et al. 2002 <sup>[[#fn:r1020|1020]]</sup> ; Scoones and Wolmer 2002 <sup>[[#fn:r1021|1021]]</sup> ). Promoting a diversity of approaches to crop–livestock interactions does not imply that the integrated model necessarily leads to land degradation, but increases the space for institutional support to local innovation (Scoones and Wolmer 2002 <sup>[[#fn:r1022|1022]]</sup> ).

However, specific managerial and technological practices that link crop and livestock production will remain an important part of the repertoire of on-farm adaptation and mitigation. Howden and coauthors (Howden et al. 2007 <sup>[[#fn:r1023|1023]]</sup> ) note the importance of innovation within existing integrated systems, including use of adapted forage crops. Rivera-Ferre et al. (2016) <sup>[[#fn:r1024|1024]]</sup> list as adaptation strategies with high potential for grazing systems, mixed crop–livestock systems or both: crop–livestock integration in general; soil management, including composting; enclosure and corralling of animals; improved storage of feed. Most of these are seen as having significant co-benefits for mitigation, and improved management of manure is seen as a mitigation measure with adaptation co-benefits.

<span id="local-and-indigenous-knowledge-for-addressing-land-degradation"></span>
=== 4.8.2 Local and indigenous knowledge for addressing land degradation ===

<div id="section-4-8-2-local-and-indigenous-knowledge-for-addressing-land-degradation-block-1"></div>

In practice, responses are anchored in scientific research, as well as local, indigenous and traditional knowledge and know-how. For example, studies in the Philippines by Camacho et al. (2016) <sup>[[#fn:r25|25]]</sup> examine how traditional integrated watershed management by indigenous people sustain regulating services vital to agricultural productivity, while delivering co-benefits in the form of biodiversity and ecosystem resilience at a landscape scale. Although responses can be site specific and sustainable at a local scale, the multi-scale interplay of drivers and pressures can nevertheless cause practices that have been sustainable for centuries to become less so. Siahaya et al. (2016) <sup>[[#fn:r1026|1026]]</sup> explore the traditional knowledge that has informed rice cultivation in the uplands of East Borneo, grounded in sophisticated shifting cultivation methods ( ''gilir balik'' ) which have been passed on for generations (more than 200 years) in order to maintain local food production. Gilir balik involves temporary cultivation of plots, after which, abandonment takes place as the land user moves to another plot, leaving the natural (forest) vegetation to return. This approach is considered sustainable if it has the support of other subsistence strategies, adapts to and integrates with the local context, and if the carrying capacity of the system is not surpassed (Siahaya et al. 2016 <sup>[[#fn:r1027|1027]]</sup> ). Often gilir balik cultivation involves intercropping of rice with bananas, cassava and other food crops. Once the abandoned plot has been left to recover such that soil fertility is restored, clearance takes place again and the plot is reused for cultivation. Rice cultivation in this way plays an important role in forest management, with several different types of succession forest being found in the study by Siahaya et al. (2016). Nevertheless, interplay of these practices with other pressures (large-scale land acquisitions for oil palm plantation, logging and mining), risk their future sustainability. Use of fire is critical in processes of land clearance, so there are also trade-offs for climate change mitigation, which have been sparsely assessed.

Interest appears to be growing in understanding how indigenous and local knowledge inform land users’ responses to degradation, as scientists engage farmers as experts in processes of knowledge co-production and co-innovation (Oliver et al. 2012 <sup>[[#fn:r1028|1028]]</sup> ; Bitzer and Bijman 2015 <sup>[[#fn:r1029|1029]]</sup> ). This can help to introduce, implement, adapt and promote the use of locally appropriate responses (Schwilch et al. 2011 <sup>[[#fn:r1030|1030]]</sup> ). Indeed, studies strongly agree on the importance of engaging local populations in both sustainable land and forest management. Meta-analyses in tropical regions that examined both forests in protected areas and community-managed forests suggest that deforestation rates are lower, with less variation in deforestation rates presenting in community-managed forests compared to protected forests (Porter-Bolland et al. 2012 <sup>[[#fn:r1031|1031]]</sup> ). This suggests that consideration of the social and economic needs of local human populations is vital in preventing forest degradation (Ward et al. 2018 <sup>[[#fn:r1032|1032]]</sup> ). However, while disciplines such as ethnopedology seek to record and understand how local people perceive, classify and use soil, and draw on that information to inform its management (Barrera-Bassols and Zinck 2003 <sup>[[#fn:r1033|1033]]</sup> ), links with climate change and its impacts (perceived and actual) are not generally considered.

<span id="reducing-deforestation-and-forest-degradation-and-increasing-afforestation"></span>
=== 4.8.3 Reducing deforestation and forest degradation and increasing afforestation ===

<div id="section-4-8-3-reducing-deforestation-and-forest-degradation-and-increasing-afforestation-block-1"></div>

Improved stewardship of forests through reduction or avoidance of deforestation and forest degradation, and enhancement of forest carbon stocks can all contribute to land-based natural climate solutions (Angelsen et al. 2018 <sup>[[#fn:r1034|1034]]</sup> ; Sonwa et al. 2011 <sup>[[#fn:r1035|1035]]</sup> ; Griscom et al. 2017 <sup>[[#fn:r1036|1036]]</sup> ). While estimates of annual emissions from tropical deforestation and forest degradation range widely from 0.5 to 3.5 GtC yr <sup>–1</sup> (Baccini et al. 2017 <sup>[[#fn:r1037|1037]]</sup> ; Houghton et al. 2012 <sup>[[#fn:r1038|1038]]</sup> ; Mitchard 2018 <sup>[[#fn:r1039|1039]]</sup> ; see also Chapter 2), they all indicate the large potential to reduce annual emissions from deforestation and forest degradation. Recent estimates of forest extent for Africa in 1900 may result in downward adjustments of historic deforestation and degradation emission estimates (Aleman et al. 2018 <sup>[[#fn:r1040|1040]]</sup> ). Emissions from forest degradation in non-Annex I countries have declined marginally from 1.1 GtCO <sub>2</sub> yr <sup>–1</sup> in 2001–2010 to 1 GtCO <sub>2</sub> yr <sup>–1</sup> in 2011–2015, but the relative emissions from degradation compared to deforestation have increased from a quarter to a third (Federici et al. 2015 <sup>[[#fn:r1041|1041]]</sup> ). Forest sector activities in developing countries were estimated to represent a technical mitigation potential in 2030 of 9 GtCO <sub>2</sub> (Miles et al. 2015). This was partitioned into reduction of deforestation (3.5 GtCO <sub>2</sub> ), reduction in degradation and forest management (1.7 GtCO <sub>2</sub> ) and afforestation and reforestation (3.8 GtCO <sub>2</sub> ). The economic mitigation potential will be lower than the technical potential (Miles et al. 2015 <sup>[[#fn:r1042|1042]]</sup> ).

Natural regeneration of second-growth forests enhances carbon sinks in the global carbon budget (Chazdon and Uriarte 2016 <sup>[[#fn:r1043|1043]]</sup> ). In Latin America, Chazdon et al. (2016) <sup>[[#fn:r1044|1044]]</sup> estimated that, in 2008, second-growth forests (up to 60 years old) covered 2.4 Mkm <sup>2</sup> of land (28.1% of the total study area). Over 40 years, these lands can potentially accumulate 8.5 GtC in above-ground biomass via low-cost natural regeneration or assisted regeneration, corresponding to a total CO <sub>2</sub> sequestration of 31.1 GtCO <sub>2</sub> (Chazdon et al. 2016b). While above-ground biomass carbon stocks are estimated to be declining in the tropics, they are increasing globally due to increasing stocks in temperate and boreal forests (Liu et al. 2015b), consistent with the observations of a global land sector carbon sink (Le Quéré et al. 2013 <sup>[[#fn:r1045|1045]]</sup> ; Keenan et al. 2017 <sup>[[#fn:r1046|1046]]</sup> ; Pan et al. 2011).

Moving from technical mitigation potentials (Miles et al. 2015 <sup>[[#fn:r1047|1047]]</sup> ) to real reduction of emissions from deforestation and forest degradation required transformational changes (Korhonen-Kurki et al. 2018 <sup>[[#fn:r1048|1048]]</sup> ). This transformation can be facilitated by two enabling conditions: the presence of already initiated policy change; or the scarcity of forest resources combined with an absence of any effective forestry framework and policies. These authors and others (Angelsen et al. 2018 <sup>[[#fn:r1049|1049]]</sup> ) found that the presence of powerful transformational coalitions of domestic pro-REDD+ (the United Nations Collaborative Programme on Reducing Emissions from Deforestation and Forest Degradation in Developing Countries) political actors combined with strong ownership and leadership, regulations and law enforcement, and performance-based funding, can provide a strong incentive for achieving REDD+ goals.

Implementing schemes such as REDD+ and various projects related to the voluntary carbon market is often regarded as a no-regrets investment (Seymour and Angelsen 2012 <sup>[[#fn:r1050|1050]]</sup> ) but the social and ecological implications (including those identified in the Cancun Safeguards) must be carefully considered for REDD+ projects to be socially and ecologically sustainable (Jagger et al. 2015 <sup>[[#fn:r1051|1051]]</sup> ). In 2018, 34 countries have submitted a REDD+ forest reference level and/ or forest reference emission level to the United Nations Framework Convention on Climate Change (UNFCCC). Of these REDD+ reference levels, 95% included the activity ‘reducing deforestation’ while 34% included the activity ‘reducing forest degradation’ (FAO 2018). Five countries submitted REDD+ results in the technical annex to their Biennial Update Report totalling an emission reduction of 6.3 GtCO <sub>2</sub> between 2006 and 2015 (FAO 2018).

Afforestation is another mitigation activity that increases carbon sequestration (Cross-Chapter Box 2 in Chapter 1). Yet, it requires careful consideration about where to plant trees to achieve potential climatic benefits, given an altering of local albedo and turbulent energy fluxes and increasing night-time land surface temperatures (Peng et al. 2014 <sup>[[#fn:r1052|1052]]</sup> ). A recent hydro-climatic modelling effort has shown that forest cover can account for about 40% of the observed decrease in annual runoff (Buendia et al. 2016 <sup>[[#fn:r1053|1053]]</sup> ). A meta-analysis of afforestation in Northern Europe (Bárcena et al. 2014 <sup>[[#fn:r1054|1054]]</sup> ) concluded that significant soil organic carbon sequestration in Northern Europe occurs after afforestation of croplands but not grasslands. Additional sequestration occurs in forest floors and biomass carbon stocks. Successful programmes of large-scale afforestation activities in South Korea and China are discussed in-depth in a special case study (Section 4.9.3).

The potential outcome of efforts to reduce emissions from deforestation and degradation in Indonesia through a 2011 moratorium on concessions to convert primary forests to either timber or palm oil uses was evaluated against rates of emissions over the period 2000 to 2010. The study concluded that less than 7% of emissions would have been avoided had the moratorium been implemented in 2000 because it only curtailed emissions due to a subset of drivers of deforestation and degradation (Busch et al. 2015 <sup>[[#fn:r1055|1055]]</sup> ).

In terms of ecological integrity of tropical forests, the policy focus on carbon storage and tree cover can be problematic if it leaves out other aspects of forests ecosystems, such as biodiversity – and particularly fauna (Panfil and Harvey 2016 <sup>[[#fn:r1056|1056]]</sup> ; Peres et al. 2016 <sup>[[#fn:r1057|1057]]</sup> ; Hinsley et al. 2015 <sup>[[#fn:r1058|1058]]</sup> ). Other concerns of forest-based projects under the voluntary carbon market are potential negative socio-economic side effects (Edstedt and Carton 2018 <sup>[[#fn:r1059|1059]]</sup> ; Carton and Andersson 2017 <sup>[[#fn:r1060|1060]]</sup> ; Osborne 2011 <sup>[[#fn:r1061|1061]]</sup> ; Scheidel and Work 2018 <sup>[[#fn:r1062|1062]]</sup> ; Richards and Lyons 2016 <sup>[[#fn:r1063|1063]]</sup> ; Borras and Franco 2018 <sup>[[#fn:r1064|1064]]</sup> ; Paladino and Fiske 2017 <sup>[[#fn:r1065|1065]]</sup> ) and leakage (particularly at the subnational scale), that is, when interventions to reduce deforestation or degradation at one site displace pressures and increase emissions elsewhere (Atmadja and Verchot 2012 <sup>[[#fn:r1066|1066]]</sup> ; Phelps et al. 2010 <sup>[[#fn:r1067|1067]]</sup> ; Lund et al. 2017 <sup>[[#fn:r1068|1068]]</sup> ; Balooni and Lund 2014 <sup>[[#fn:r1069|1069]]</sup> ).

Maintaining and increasing forest area, in particular native forests rather than monoculture and short-rotation plantations, contributes to the maintenance of global forest carbon stocks (Lewis et al. 2019 <sup>[[#fn:r1070|1070]]</sup> ) ( ''robust evidence, high agreement'' ).

<span id="sustainable-forest-management-sfm-and-co2-removal-cdr-technologies"></span>
=== 4.8.4 Sustainable forest management (SFM) and CO2 removal (CDR) technologies ===

<div id="section-4-8-4-sustainable-forest-management-sfm-and-co2-removal-cdr-technologies-block-1"></div>

While reducing deforestation and forest degradation may directly help to meet mitigation goals, SFM aimed at providing timber, fibre, biomass and non-timber resources can provide long-term livelihood for communities, reduce the risk of forest conversion to non-forest uses (settlement, crops, etc.), and maintain land productivity, thus reducing the risks of land degradation (Putz et al. 2012 <sup>[[#fn:r1071|1071]]</sup> ; Gideon Neba et al. 2014 <sup>[[#fn:r1072|1072]]</sup> ; Sufo Kankeu et al. 2016 <sup>[[#fn:r1073|1073]]</sup> ; Nitcheu Tchiadje et al. 2016 <sup>[[#fn:r1074|1074]]</sup> ; Rossi et al. 2017 <sup>[[#fn:r1075|1075]]</sup> ).

Developing SFM strategies aimed at contributing towards negative emissions throughout this century requires an understanding of forest management impacts on ecosystem carbon stocks (including soils), carbon sinks, carbon fluxes in harvested wood, carbon storage in harvested wood products, including landfills and the emission reductions achieved through the use of wood products and bioenergy (Nabuurs et al. 2007 <sup>[[#fn:r1076|1076]]</sup> ; Lemprière et al. 2013 <sup>[[#fn:r1077|1077]]</sup> ; Kurz et al. 2016 <sup>[[#fn:r1078|1078]]</sup> ; Law et al. 2018 <sup>[[#fn:r1079|1079]]</sup> ; Nabuurs et al. 2017 <sup>[[#fn:r1080|1080]]</sup> ). Transitions from natural to managed forest landscapes can involve a reduction in forest carbon stocks, the magnitude of which depends on the initial landscape conditions, the harvest rotation length relative to the frequency and intensity of natural disturbances, and on the age-dependence of managed and natural disturbances (Harmon et al. 1990 <sup>[[#fn:r1081|1081]]</sup> ; Kurz et al. 1998 <sup>[[#fn:r1082|1082]]</sup> ). Initial landscape conditions, in particular the age-class distribution and therefore carbon stocks of the landscape, strongly affect the mitigation potential of forest management options (Ter-Mikaelian et al. 2013 <sup>[[#fn:r1083|1083]]</sup> ; Kilpeläinen et al. 2017 <sup>[[#fn:r1084|1084]]</sup> ). Landscapes with predominantly mature forests may experience larger reductions in carbon stocks during the transition to managed landscapes (Harmon et al. 1990 <sup>[[#fn:r1085|1085]]</sup> ; Kurz et al. 1998 <sup>[[#fn:r1086|1086]]</sup> ; Lewis et al. 2019 <sup>[[#fn:r1087|1087]]</sup> ). In landscapes with predominantly young or recently disturbed forests, SFM can enhance carbon stocks (Henttonen et al. 2017 <sup>[[#fn:r1088|1088]]</sup> ).

Forest growth rates, net primary productivity, and net ecosystem productivity are age-dependent, with maximum rates of CO <sub>2</sub> removal (CDR) from the atmosphere occurring in young to medium-aged forests and declining thereafter (Tang et al. 2014 <sup>[[#fn:r1089|1089]]</sup> ). In boreal forest ecosystem, estimation of carbon stocks and carbon fluxes indicate that old growth stands are typically small carbon sinks or carbon sources (Gao et al. 2018 <sup>[[#fn:r1090|1090]]</sup> ; Taylor et al. 2014 <sup>[[#fn:r1091|1091]]</sup> ; Hadden and Grelle 2016 <sup>[[#fn:r1092|1092]]</sup> ). In tropical forests, carbon uptake rates in the first 20 years of forest recovery were 11 times higher than uptake rates in old-growth forests (Poorter et al. 2016 <sup>[[#fn:r1093|1093]]</sup> ). Age-dependent increases in forest carbon stocks and declines in forest carbon sinks mean that landscapes with older forests have accumulated more carbon but their sink strength is diminishing, while landscapes with younger forests contain less carbon but they are removing CO <sub>2</sub> from the atmosphere at a much higher rate (Volkova et al. 2017 <sup>[[#fn:r1094|1094]]</sup> ; Poorter et al. 2016 <sup>[[#fn:r1095|1095]]</sup> ). The rates of CDR are not just age-related but also controlled by many biophysical factors and human activities (Bernal et al. 2018 <sup>[[#fn:r1096|1096]]</sup> ). In ecosystems with uneven-aged, multispecies forests, the relationships between carbon stocks and sinks are more difficult and expensive to quantify.

Whether or not forest harvest and use of biomass is contributing to net reductions of atmospheric carbon depends on carbon losses during and following harvest, rates of forest regrowth, and the use of harvested wood and carbon retention in long-lived or short-lived products, as well as the emission reductions achieved through the substitution of emissions-intensive products with wood products (Lemprière et al. 2013 <sup>[[#fn:r1097|1097]]</sup> ; Lundmark et al. 2014 <sup>[[#fn:r1098|1098]]</sup> ; Xu et al. 2018b <sup>[[#fn:r1099|1099]]</sup> ; Olguin et al. 2018 <sup>[[#fn:r1100|1100]]</sup> ; Dugan et al. 2018 <sup>[[#fn:r1101|1101]]</sup> ; Chen et al. 2018b <sup>[[#fn:r1102|1102]]</sup> ; Pingoud et al. 2018 <sup>[[#fn:r1103|1103]]</sup> ; Seidl et al. 2007 <sup>[[#fn:r1104|1104]]</sup> ). Studies that ignore changes in forest carbon stocks (such as some lifecycle analyses that assume no impacts of harvest on forest carbon stocks), ignore changes in wood product pools (Mackey et al. 2013 <sup>[[#fn:r1105|1105]]</sup> ) or assume long-term steady state (Pingoud et al. 2018 <sup>[[#fn:r1106|1106]]</sup> ), or ignore changes in emissions from substitution benefits (Mackey et al. 2013 <sup>[[#fn:r1107|1107]]</sup> ; Lewis et al. 2019 <sup>[[#fn:r1108|1108]]</sup> ) will arrive at diverging conclusions about the benefits of SFM. Moreover, assessments of climate benefits of any mitigation action must also consider the time dynamics of atmospheric impacts, as some actions will have immediate benefits (e.g., avoided deforestation), while others may not achieve net atmospheric benefits for decades or centuries. For example, the climate benefits of woody biomass use for bioenergy depend on several factors, such as the source and alternate fate of the biomass, the energy type it substitutes, and the rates of regrowth of the harvested forest (Laganière et al. 2017 <sup>[[#fn:r1109|1109]]</sup> ; Ter-Mikaelian et al. 2014 <sup>[[#fn:r1110|1110]]</sup> ; Smyth et al. 2017 <sup>[[#fn:r1111|1111]]</sup> ). Conversion of primary forests in regions of very low stand-replacing disturbances to short-rotation plantations where the harvested wood is used for short-lived products with low displacement factors will increase emissions. In general, greater mitigation benefits are achieved if harvested wood products are used for products with long carbon retention time and high displacement factors.

With increasing forest age, carbon sinks in forests will diminish until harvest or natural disturbances, such as wildfire, remove biomass carbon or release it to the atmosphere (Seidl et al. 2017 <sup>[[#fn:r1112|1112]]</sup> ). While individual trees can accumulate carbon for centuries (Köhl et al. 2017 <sup>[[#fn:r1113|1113]]</sup> ), stand-level carbon accumulation rates depend on both tree growth and tree mortality rates (Hember et al. 2016 <sup>[[#fn:r1114|1114]]</sup> ; Lewis et al. 2004 <sup>[[#fn:r1115|1115]]</sup> ). SFM, including harvest and forest regeneration, can help maintain active carbon sinks by maintaining a forest age-class distribution that includes a share of young, actively growing stands (Volkova et al. 2018 <sup>[[#fn:r1116|1116]]</sup> ; Nabuurs et al. 2017 <sup>[[#fn:r1117|1117]]</sup> ). The use of the harvested carbon in either long-lived wood products (e.g., for construction), short-lived wood products (e.g., pulp and paper), or biofuels affects the net carbon balance of the forest sector (Lemprière et al. 2013 <sup>[[#fn:r1118|1118]]</sup> ; Matthews et al. 2018 <sup>[[#fn:r1119|1119]]</sup> ). The use of these wood products can further contribute to GHG emission-reduction goals by avoiding the emissions from the products with higher embodied emissions that have been displaced (Nabuurs et al. 2007 <sup>[[#fn:r1120|1120]]</sup> ; Lemprière et al. 2013 <sup>[[#fn:r1121|1121]]</sup> ). In 2007 the IPCC concluded that ‘[i]n the long term, a sustainable forest management strategy aimed at maintaining or increasing forest carbon stocks, while producing an annual sustained yield of timber, fibre or energy from the forest, will generate the largest sustained mitigation benefit’ (Nabuurs et al. 2007 <sup>[[#fn:r1122|1122]]</sup> ). The apparent trade-offs between maximising forest carbon stocks and maximising ecosystem carbon sinks are at the origin of ongoing debates about optimum management strategies to achieve negative emissions (Keith et al. 2014 <sup>[[#fn:r1123|1123]]</sup> ; Kurz et al. 2016 <sup>[[#fn:r1124|1124]]</sup> ; Lundmark et al. 2014 <sup>[[#fn:r1125|1125]]</sup> ). SFM, including the intensification of carbon-focused management strategies, can make long-term contributions towards negative emissions if the sustainability of management is assured through appropriate governance, monitoring and enforcement. As specified in the definition of SFM, other criteria such as biodiversity must also be considered when assessing mitigation outcomes (Lecina-Diaz et al. 2018 <sup>[[#fn:r1126|1126]]</sup> ). Moreover, the impacts of changes in management on albedo and other non-GHG factors also need to be considered (Luyssaert et al. 2018 <sup>[[#fn:r1127|1127]]</sup> ) (Chapter 2). The contribution of SFM for negative emissions is strongly affected by the use of the wood products derived from forest harvest and the time horizon over which the carbon balance is assessed. SFM needs to anticipate the impacts of climate change on future tree growth, mortality and disturbances when designing climate change mitigation and adaptation strategies (Valade et al. 2017 <sup>[[#fn:r1128|1128]]</sup> ; Seidl et al. 2017 <sup>[[#fn:r1129|1129]]</sup> ).

<span id="policy-responses-to-land-degradation"></span>
=== 4.8.5 Policy responses to land degradation ===

<div id="section-4-8-5-policy-responses-to-land-degradation-block-1"></div>

The 1992 United Nations Conference on Environment and Development (UNCED), also known as the Rio de Janeiro Earth Summit, recognised land degradation as a major challenge to sustainable development, and led to the establishment of the UNCCD, which specifically addressed land degradation in the drylands. The UNCCD emphasises sustainable land use to link poverty reduction on one hand and environmental protection on the other. The two other ‘Rio Conventions’ emerging from the UNCED – the UNFCCC and the Convention on Biological Diversity (CBD) – focus on climate change and biodiversity, respectively. The land has been recognised as an aspect of common interest to the three conventions, and SLM is proposed as a unifying theme for current global efforts on combating land degradation, climate change and loss of biodiversity, as well as facilitating land-based adaptation to climate change and sustainable development.

The Global Environmental Facility (GEF) funds developing countries to undertake activities that meet the goals of the conventions and deliver global environmental benefits. Since 2002, the GEF has invested in projects that support SLM through its Land Degradation Focal Area Strategy, to address land degradation within and beyond the drylands.

Under the UNFCCC, parties have devised National Adaptation Plans (NAPs) that identify medium- and long-term adaptation needs. Parties have also developed their climate change mitigation plans, presented as NDCs. These programmes have the potential of assisting the promotion of SLM. It is understood that the root causes of land degradation and successful adaptation will not be realised until holistic solutions to land management are explored. SLM can help address root causes of low productivity, land degradation, loss of income-generating capacity, as well as contribute to the amelioration of the adverse effects of climate change.

The ‘4 per 1000’ (4p1000) initiative (Soussana et al. 2019 <sup>[[#fn:r1130|1130]]</sup> ) launched by France during the UNFCCC COP21 in 2015 aims at capturing CO <sub>2</sub> from the atmosphere through changes to agricultural and forestry practices at a rate that would increase the carbon content of soils by 0.4% per year (Rumpel et al. 2018 <sup>[[#fn:r1131|1131]]</sup> ). If global soil carbon content increases at this rate in the top 30–40 cm, the annual increase in atmospheric CO <sub>2</sub> would be stopped (Dignac et al. 2017 <sup>[[#fn:r1132|1132]]</sup> ). This is an illustration of how extremely important soils are for addressing climate change. The initiative is based on eight steps: stop carbon loss (priority #1 is peat soils); promote carbon uptake; monitor, report and verify impacts; deploy technology for tracking soil carbon; test strategies for implementation and upscaling; involve communities; coordinate policies; and provide support (Rumpel et al. 2018 <sup>[[#fn:r1133|1133]]</sup> ). Questions remain, however, about the extent that the 4p1000 is achievable as a universal goal (van Groenigen et al. 2017 <sup>[[#fn:r1134|1134]]</sup> ; Poulton et al. 2018 <sup>[[#fn:r1135|1135]]</sup> ; Schlesinger and Amundson 2018 <sup>[[#fn:r1136|1136]]</sup> ).

LDN was introduced by the UNCCD at Rio +20, and adopted at UNCCD COP12 (UNCCD 2016a <sup>[[#fn:r1137|1137]]</sup> ). LDN is defined as ‘a state whereby the amount and quality of land resources necessary to support ecosystem functions and services and enhance food security remain stable or increase within specified temporal and spatial scales and ecosystems’(Cowie et al. 2018 <sup>[[#fn:r1138|1138]]</sup> ). Pursuit of LDN requires effort to avoid further net loss of the land-based natural capital relative to a reference state, or baseline. LDN encourages a dual-pronged effort involving SLM to reduce the risk of land degradation, combined with efforts in land restoration and rehabilitation, to maintain or enhance land-based natural capital, and its associated ecosystem services (Orr et al. 2017 <sup>[[#fn:r1139|1139]]</sup> ; Cowie et al. 2018 <sup>[[#fn:r1140|1140]]</sup> ). Planning for LDN involves projecting the expected cumulative impacts of land-use and land management decisions, then counterbalancing anticipated losses with measures to achieve equivalent gains, within individual land types (where land type is defined by land potential). Under the LDN framework developed by UNCCD, three primary indicators are used to assess whether LDN is achieved by 2030: land cover change; net primary productivity; and soil organic carbon (Cowie et al. 2018 <sup>[[#fn:r1141|1141]]</sup> ; Sims et al. 2019 <sup>[[#fn:r1142|1142]]</sup> ). Achieving LDN therefore requires integrated landscape management that seeks to optimise land use to meet multiple objectives (ecosystem health, food security, human well-being) (Cohen-Shacham et al. 2016 <sup>[[#fn:r1143|1143]]</sup> ). The response hierarchy of Avoid &gt; Reduce &gt; Reverse land degradation articulates the priorities in planning LDN interventions. LDN provides the impetus for widespread adoption of SLM and efforts to restore or rehabilitate land. Through its focus, LDN ultimately provides tremendous potential for mitigation of, and adaptation to, climate change by halting and reversing land degradation and transforming land from a carbon source to a sink. There are strong synergies between the concept of LDN and the NDCs of many countries, with linkages to national climate plans. LDN is also closely related to many Sustainable Development Goals (SDG) in the areas of poverty, food security, environmental protection and sustainable use of natural resources (UNCCD 2016b <sup>[[#fn:r1144|1144]]</sup> ). The GEF is supporting countries to set LDN targets and implement their LDN plans through its land degradation focal area, which encourages application of integrated landscape approaches to managing land degradation (GEF 2018 <sup>[[#fn:r1145|1145]]</sup> ).

The 2030 Agenda for Sustainable Development, adopted by the United Nations in 2015, comprises 17 SDGs. Goal 15 is of direct relevance to land degradation, with the objective to protect, restore and promote sustainable use of terrestrial ecosystems, sustainably manage forests, combat desertification and halt and reverse land degradation and halt biodiversity loss. Target 15.3 specifically addresses LDN. Other goals that are relevant for land degradation include Goal 2 (Zero hunger), Goal 3 (Good health and well-being), Goal 7 (Affordable and clean energy), Goal 11 (Sustainable cities and communities), and Goal 12 (Responsible production and consumption). Sustainable management of land resources underpins the SDGs related to hunger, climate change and environment. Further goals of a cross-cutting nature include 1 (No poverty), 6 (Clean water and sanitation) and 13 (Climate action). It remains to be seen how these interconnections are dealt with in practice.

With a focus on biodiversity, IPBES published a comprehensive assessment of land degradation in 2018 (Montanarella et al. 2018 <sup>[[#fn:r1146|1146]]</sup> ). The IPBES report, together with this report focusing on climate change, may contribute to creating a synergy between the two main global challenges for addressing land degradation in order to help achieve the targets of SDG 15 (protect, restore and promote sustainable use of terrestrial ecosystems, sustainably manage forests, combat desertification, and halt and reverse land degradation and halt biodiversity loss).

Market-based mechanisms like the Clean Development Mechanism (CDM) under the UNFCCC and the voluntary carbon market provide incentives to enhance carbon sinks on the land through afforestation and reforestation. Implications for local land use and food security have been raised as a concern and need to be assessed (Edstedt and Carton 2018 <sup>[[#fn:r1147|1147]]</sup> ; Olsson et al. 2014b <sup>[[#fn:r1148|1148]]</sup> ). Many projects aimed at reducing emissions from deforestation and forest degradations (not to be confused with the national REDD+ programmes in accordance with the UNFCCC Warsaw Framework) are being planned and implemented to primarily target countries with high forest cover and high deforestation rates. Some parameters of incentivising emissions reduction, quality of forest governance, conservation priorities, local rights and tenure frameworks, and sub-national project potential are being looked into, with often very mixed results (Newton et al. 2016 <sup>[[#fn:r1149|1149]]</sup> ; Gebara and Agrawal 2017 <sup>[[#fn:r1150|1150]]</sup> ).

Besides international public initiatives, some actors in the private sector are increasingly aware of the negative environmental impacts of some global value chains producing food, fibre, and energy products (Lambin et al. 2018 <sup>[[#fn:r1151|1151]]</sup> ; van der Ven and Cashore 2018 <sup>[[#fn:r1152|1152]]</sup> ; van der Ven et al. 2018 <sup>[[#fn:r1153|1153]]</sup> ; Lyons-White and Knight 2018 <sup>[[#fn:r1154|1154]]</sup> ). While improvements are underway in many supply chains, measures implemented so far are often insufficient to be effective in reducing or stopping deforestation and forest degradation (Lambin et al. 2018 <sup>[[#fn:r1155|1155]]</sup> ). The GEF is investing in actions to reduce deforestation in commodity supply chains through its Food Systems, Land Use, and Restoration Impact Program (GEF 2018 <sup>[[#fn:r1156|1156]]</sup> ).

<div id="section-4-8-5-1-limits-to-adaptation"></div>

<span id="limits-to-adaptation"></span>
==== 4.8.5.1 Limits to adaptation ====

<div id="section-4-8-5-1-limits-to-adaptation-block-1"></div>

SLM can be deployed as a powerful adaptation strategy in most instances of climate change impacts on natural and social systems, yet there are limits to adaptation (Klein et al. 2014 <sup>[[#fn:r1157|1157]]</sup> ; Dow, Berhout and Preston 2013 <sup>[[#fn:r1158|1158]]</sup> ). Such limits are dynamic and interact with social and institutional conditions (Barnett et al. 2015 <sup>[[#fn:r1159|1159]]</sup> ; Filho and Nalau 2018 <sup>[[#fn:r1160|1160]]</sup> ). Exceeding adaptation limits will trigger escalating losses or require undesirable transformational change, such as forced migration. The rate of change in relation to the rate of possible adaptation is crucial (Dow et al. 2013 <sup>[[#fn:r1161|1161]]</sup> ). How limits to adaptation are defined, and how they can be measured, is contextual and contested. Limits must be assessed in relation to the ultimate goals of adaptation, which is subject to diverse and differential values (Dow et al. 2013 <sup>[[#fn:r1162|1162]]</sup> ; Adger et al. 2009 <sup>[[#fn:r1163|1163]]</sup> ). A particularly sensitive issue is whether migration is accepted as adaptation or not (Black et al. 2011 <sup>[[#fn:r1164|1164]]</sup> ; Tacoli 2009 <sup>[[#fn:r1165|1165]]</sup> ; Bardsley and Hugo 2010 <sup>[[#fn:r1166|1166]]</sup> ). If migration were understood and accepted as a form of successful adaptation, it would change the limits to adaptation by reducing, or even avoiding, future humanitarian crises caused by climate extremes (Adger et al. 2009 <sup>[[#fn:r1167|1167]]</sup> ; Upadhyay et al. 2017 <sup>[[#fn:r1168|1168]]</sup> ; Nalau et al. 2018 <sup>[[#fn:r1169|1169]]</sup> ).

In the context of land degradation, potential limits to adaptation exist if land degradation becomes so severe and irreversible that livelihoods cannot be maintained, and if migration is either not acceptable or not possible. Examples are coastal erosion where land disappears (Gharbaoui and Blocher 2016 <sup>[[#fn:r1170|1170]]</sup> ; Luetz 2018 <sup>[[#fn:r1171|1171]]</sup> ), collapsing livelihoods due to thawing of permafrost (Landauer and Juhola 2019 <sup>[[#fn:r1172|1172]]</sup> ), and extreme forms of soil erosion, (e.g., landslides (Van der Geest and Schindler 2016 <sup>[[#fn:r1173|1173]]</sup> ) and gully erosion leading to badlands (Poesen et al. 2003 <sup>[[#fn:r1174|1174]]</sup> )).

<span id="resilience-and-thresholds"></span>
=== 4.8.6 Resilience and thresholds ===

<div id="section-4-8-6-resilience-and-thresholds-block-1"></div>

Resilience refers to the capacity of interconnected social, economic and ecological systems, such as farming systems, to absorb disturbance (e.g., drought, conflict, market collapse), and respond or reorganise, to maintain their essential function, identity and structure. Resilience can be described as ‘coping capacity’. The disturbance may be a shock – sudden events such as a flood or disease epidemic – or it may be a trend that develops slowly, like a drought or market shift. The shocks and trends anticipated to occur due to climate change are expected to exacerbate risk of land degradation. Therefore, assessing and enhancing resilience to climate change is a critical component of designing SLM strategies.

Resilience as an analytical lens is particularly strong in ecology and related research on natural resource management (Folke et al. 2010 <sup>[[#fn:r1175|1175]]</sup> ; Quinlan et al. 2016 <sup>[[#fn:r1176|1176]]</sup> ) while, in the social sciences, the relevance of resilience for studying social and ecological interactions is contested

(Cote and Nightingale 2012 <sup>[[#fn:r1177|1177]]</sup> ; Olsson et al. 2015 <sup>[[#fn:r1178|1178]]</sup> ; Cretney 2014 <sup>[[#fn:r1179|1179]]</sup> ; Béné et al. 2012 <sup>[[#fn:r1180|1180]]</sup> ; Joseph 2013 <sup>[[#fn:r1181|1181]]</sup> ). In the case of adaptation to climate change (and particularly regarding limits to adaptation), a crucial ambiguity of resilience is the question of whether resilience is a normative concept (i.e., resilience is good or bad) or a descriptive characteristic of a system (i.e., neither good nor bad). Previous IPCC reports have defined resilience as a normative (positive) attribute (see AR5 Glossary), while the wider scientific literature is divided on this (Weichselgartner and Kelman 2015 <sup>[[#fn:r1182|1182]]</sup> ; Strunz 2012 <sup>[[#fn:r1183|1183]]</sup> ; Brown 2014 <sup>[[#fn:r1184|1184]]</sup> ; Grimm and Calabrese 2011 <sup>[[#fn:r1185|1185]]</sup> ; Thorén and Olsson 2018 <sup>[[#fn:r1186|1186]]</sup> ). For example, is outmigration from a disaster-prone area considered a successful adaptation (high resilience) or a collapse of the livelihood system (lack of resilience) (Thorén and Olsson 2018 <sup>[[#fn:r1187|1187]]</sup> )? In this report, resilience is considered a positive attribute when it maintains capacity for adaptation, learning and/or transformation.

Furthermore, ‘resilience’ and the related terms ‘adaptation’ and ‘transformation’ are defined and used differently by different communities (Quinlan et al. 2016 <sup>[[#fn:r1188|1188]]</sup> ). The relationship and hierarchy of resilience with respect to vulnerability and adaptive capacity are also debated, with different perspectives between disaster management and global change communities, (e.g., Cutter et al. 2008 <sup>[[#fn:r1189|1189]]</sup> ). Nevertheless, these differences in usage need not inhibit the application of ‘resilience thinking’ in managing land degradation; researchers using these terms, despite variation in definitions, apply the same fundamental concepts to inform management of human-environment systems, to maintain or improve the resource base, and sustain livelihoods.

Applying resilience concepts involves viewing the land as a component of an interlinked social-ecological system; identifying key relationships that determine system function and vulnerabilities of the system; identifying thresholds or tipping points beyond which the system transitions to an undesirable state; and devising management strategies to steer away from thresholds of potential concern, thus facilitating healthy systems and sustainable production (Walker et al. 2009 <sup>[[#fn:r1190|1190]]</sup> ).

A threshold is a non-linearity between a controlling variable and system function, such that a small change in the variable causes the system to shift to an alternative state. Bestelmeyer et al. (2015) <sup>[[#fn:r1191|1191]]</sup> and Prince et al. (2018) <sup>[[#fn:r1192|1192]]</sup> illustrate this concept in the context of land degradation. Studies have identified various biophysical and socio-economic thresholds in different land-use systems. For example, 50% ground cover (living and dead plant material and biological crusts) is a recognised threshold for dryland grazing systems (e.g., Tighe et al. 2012 <sup>[[#fn:r1193|1193]]</sup> ); below this threshold, the infiltration rate declines, risk of erosion causing loss of topsoil increases, a switch from perennial to annual grass species occurs and there is a consequential sharp decline in productivity. This shift to a lower-productivity state cannot be reversed without significant human intervention. Similarly, the combined pressure of water limitations and frequent fire can lead to transition from closed forest to savannah or grassland: if fire is too frequent, trees do not reach reproductive maturity and post-fire regeneration will fail; likewise, reduced rainfall/increased drought prevents successful forest regeneration (Reyer et al. 2015 <sup>[[#fn:r1194|1194]]</sup> ; Thompson et al. 2009 <sup>[[#fn:r1195|1195]]</sup> ) (Cross-Chapter Box 3 in Chapter 2).

In managing land degradation, it is important to assess the resilience of the existing system, and the proposed management interventions. If the existing system is in an undesirable state or considered unviable under expected climate trends, it may be desirable to promote adaptation or even transformation to a different system that is more resilient to future changes. For example, in an irrigation district where water shortages are predicted, measures could be implemented to improve water use efficiency, for example, by establishing drip irrigation systems for water delivery, although transformation to pastoralism or mixed dryland cropping/livestock production may be more sustainable in the longer term, at least for part of the area. Application of SLM practices, especially those focused on ecological functions (e.g., agroecology, ecosystem-based approaches, regenerative agriculture, organic farming), can be effective in building resilience of agro-ecosystems (Henry et al. 2018). Similarly, the resilience of managed forests can be enhanced by SFM that protects or enhances biodiversity, including assisted migration of tree species within their current range limit (Winder et al. 2011 <sup>[[#fn:r1197|1197]]</sup> ; Pedlar et al. 2012 <sup>[[#fn:r1198|1198]]</sup> ) or increasing species diversity in plantation forests (Felton et al. 2010 <sup>[[#fn:r1199|1199]]</sup> ; Liu et al. 2018a <sup>[[#fn:r1200|1200]]</sup> ). The essential features of a resilience approach to management of land degradation under climate change are described by O’Connell et al. (2016) <sup>[[#fn:r1201|1201]]</sup> and Simonsen et al. (2014) <sup>[[#fn:r1202|1202]]</sup> .

Consideration of resilience can enhance effectiveness of interventions to reduce or reverse land degradation ( ''medium agreement, limited evidence'' ). This approach will increase the likelihood that SLM/SFM and land restoration/rehabilitation interventions achieve long-term environmental and social benefits. Thus, consideration of resilience concepts can enhance the capacity of land systems to cope with climate change and resist land degradation, and assist land-use systems to adapt to climate change.

<span id="barriers-to-implementation-of-sustainable-land-management-slm"></span>
=== 4.8.7 Barriers to implementation of sustainable land management (SLM) ===

<div id="section-4-8-7-barriers-to-implementation-of-sustainable-land-management-slm-block-1"></div>

There is a growing recognition that addressing barriers and designing solutions to complex environmental problems, such as land degradation, requires awareness of the larger system into which the problems and solutions are embedded (Laniak et al. 2013 <sup>[[#fn:r1203|1203]]</sup> ). An ecosystem approach to sustainable land management (SLM) based on an understanding of land degradation processes has been recommended to separate multiple drivers, pressures and impacts (Kassam et al. 2013 <sup>[[#fn:r1204|1204]]</sup> ), but large uncertainty in model projections of future climate, and associated ecosystem processes (IPCC 2013a <sup>[[#fn:r1205|1205]]</sup> ) pose additional challenges to the implementation of SLM. As discussed earlier in this chapter, many SLM practices, including technologies and approaches, are available that can increase yields and contribute to closing the yield gap between actual and potential crop or pasture yield, while also enhancing resilience to climate change (Yengoh and Ardö 2014 <sup>[[#fn:r1206|1206]]</sup> ; WOCAT n.d.). However, there are often systemic barriers to adoption and scaling up of SLM practices, especially in developing countries.

Uitto (2016) <sup>[[#fn:r1207|1207]]</sup> identified areas that the GEF, the financial mechanism of the UNCCD, UNFCCC and other multilateral environmental agreements, can address to solve global environmental problems. These include: removal of barriers related to knowledge and information; strategies for implementation of technologies and approaches; and institutional capacity. Strengthening these areas would drive transformational change, leading to behavioural change and broader adoption of sustainable environmental practices. Detailed analysis of barriers as well as strategies, methods and approaches to scale up SLM have been undertaken for GEF programmes in Africa, China and globally (Tengberg and Valencia 2018 <sup>[[#fn:r1208|1208]]</sup> ; Liniger et al. 2011 <sup>[[#fn:r1209|1209]]</sup> ; Tengberg et al. 2016 <sup>[[#fn:r1210|1210]]</sup> ). A number of interconnected barriers and bottlenecks to the scaling up of SLM have been identified in this context and are related to:

* limited access to knowledge and information, including new SLM technologies and problem-solving capacities
* weak enabling environment, including the policy, institutional and legal framework for SLM, and land tenure and property rights
* inadequate learning and adaptive knowledge management in the project cycle, including monitoring and evaluation of impacts
* limited access to finance for scaling up, including public and private funding, innovative business models for SLM technologies and financial mechanisms and incentives, such as payment for ecosystem services (PES), insurance and micro-credit schemes(see also Shames et al. 2014).Adoption of innovations and new technologies are increasingly analysed using the transition theory framework (Geels 2002 <sup>[[#fn:r1211|1211]]</sup> ), the starting point being the recognition that many global environmental problems cannot be solved by technological change alone, but require more far-reaching change of social-ecological systems. Using transition theory makes it possible to analyse how adoption and implementation follow the four stages of sociotechnical transitions,

from predevelopment of technologies and approaches at the niche level, take-off and acceleration, to regime shift and stabilisation at the landscape level. According to a recent review of sustainability transitions in developing countries (Wieczorek 2018 <sup>[[#fn:r1212|1212]]</sup> ), three internal niche processes are important, including the formation of networks that support and nurture innovation, the learning process, and the articulation of expectations to guide the learning process. While technologies are important, institutional and political aspects form the major barriers to transition and upscaling. In developing and transition economies, informal institutions play a pivotal role, and transnational linkages are also important, such as global value chains. In these countries, it is therefore more difficult to establish fully coherent regimes or groups of individuals who share expectations, beliefs or behaviour, as there is a high level of uncertainty about rules and social networks or dominance of informal institutions, which creates barriers to change. This uncertainty is further exacerbated by climate change. Landscape forces comprise a set of slow-changing factors, such as broad cultural and normative values, long-term economic effects such as urbanisation, and shocks such as war and crises that can lead to change.

A study on SLM in the Kenyan highlands using transition theory concluded that barriers to adoption of SLM included high poverty levels, a low-input/low-output farming system with limited potential to generate income, diminishing land sizes, and low involvement of the youth in farming activities. Coupled with a poor coordination of government policies for agriculture and forestry, these barriers created negative feedbacks in the SLM transition process. Other factors to consider include gender issues and lack of secure land tenure. Scaling up of SLM technologies would require collaboration of diverse stakeholders across multiple scales, a more supportive policy environment and substantial resource mobilisation (Mutoko et al. 2014 <sup>[[#fn:r1213|1213]]</sup> ). Tengberg and Valencia (2018) <sup>[[#fn:r1214|1214]]</sup> analysed the findings from a review of the GEF’s integrated natural resources management portfolio of projects using the transition theory framework (Figure 4.7).

<div id="section-4-8-7-barriers-to-implementation-of-sustainable-land-management-slm-block-2"></div>

<span id="figure-4.7"></span>
<!-- START IMG -->
<!-- IMG TITLE -->
'''Figure 4.7'''

<span id="the-transition-from-slm-niche-adoption-to-regime-shift-and-landscape-development.-figure-draws-inspiration-from-geels-2002-adapted-from-tengberg-and-valencia-2018."></span>
<!-- IMG CAPTION -->
'''The transition from SLM niche adoption to regime shift and landscape development. Figure draws inspiration from Geels (2002), adapted from Tengberg and Valencia (2018).'''
<!-- IMG FILE -->
[[File:08e4e20f15b763ef30482c26a712361f Figure-4.7.jpg]]

The transition from SLM niche adoption to regime shift and landscape development. Figure draws inspiration from Geels (2002) <sup>[[#fn:r1653|1653]]</sup> , adapted from Tengberg and Valencia (2018) <sup>[[#fn:r1654|1654]]</sup> .

<!-- END IMG -->
<div id="section-4-8-7-barriers-to-implementation-of-sustainable-land-management-slm-block-3"></div>

They concluded that to remove barriers to SLM, an agricultural innovations systems approach that supports co-production of knowledge with multiple stakeholders, institutional innovations, a focus on value chains and strengthening of social capital to facilitate shared learning and collaboration could accelerate the scaling up of sustainable technologies and practices from the niche to the landscape level. Policy integration and establishment of financial mechanisms and incentives could contribute to overcoming barriers to a regime shift. The new SLM regime could, in turn, be stabilised and sustained at the landscape level by multi-stakeholder knowledge platforms and strategic partnerships. However, transitions to more sustainable regimes and practices are often challenged by lock-in mechanisms in the current system (Lawhon and Murphy 2012 <sup>[[#fn:r1215|1215]]</sup> ) such as economies of scale, investments already made in equipment, infrastructure and competencies, lobbying, shared beliefs, and practices, that could hamper wider adoption of SLM.

Adaptive, multi-level and participatory governance of social-ecological systems is considered important for regime shifts and transitions to take place (Wieczorek 2018 <sup>[[#fn:r1216|1216]]</sup> ) and essential to secure the capacity of environmental assets to support societal development over longer time periods (Folke et al. 2005 <sup>[[#fn:r1217|1217]]</sup> ). There is also recognition that effective environmental policies and programmes need to be informed by a comprehensive understanding of the biophysical, social, and economic components and processes of a system, their complex interactions, and how they respond to different changes (Kelly (Letcher) et al. 2013). But blueprint policies will not work, due to the wide diversity of rules and informal institutions used across sectors and regions of the world, especially in traditional societies (Ostrom 2009 <sup>[[#fn:r1218|1218]]</sup> ).

The most effective way of removing barriers to funding of SLM has been mainstreaming of SLM objectives and priorities into relevant policy and development frameworks, and combining SLM best practices with economic incentives for land users. As the short-term costs for establishing and maintaining SLM measures are generally high and constitute a barrier to adoption, land users may need to be compensated for generation of longer-term public goods, such as ecosystem services. Cost-benefit analyses can be conducted on SLM interventions to facilitate such compensations (Liniger et al. 2011 <sup>[[#fn:r1219|1219]]</sup> ; Nkonya et al. 2016 <sup>[[#fn:r1220|1220]]</sup> ; Tengberg et al. 2016 <sup>[[#fn:r1221|1221]]</sup> ). The landscape approach is a means to reconcile competing demands on the land and remove barriers to implementation of SLM (e.g., Sayer et al. 2013 <sup>[[#fn:r1222|1222]]</sup> ; Bürgi et al. 2017 <sup>[[#fn:r1223|1223]]</sup> ). It involves an increased focus on participatory governance, development of new SLM business models, and innovative funding schemes, including insurance (Shames et al. 2014 <sup>[[#fn:r1224|1224]]</sup> ). The LDN Fund takes a landscape approach and raises private finance for SLM and promotes market-based instruments, such as PES, certification and carbon trading, that can support scaling up of SLM to improve local livelihoods, sequester carbon and enhance the resilience to climate change (Baumber et al. 2019 <sup>[[#fn:r1225|1225]]</sup> ).

<span id="case-studies"></span>