Editing Test:SRCCL/Chapter-1 (section)

== CCB1 Scenarios and other methods to characterise the future of land ==

<div id="section-1-2-2-2-nature-and-scope-of-uncertainties-related-to-land-use-block-1"></div>

Mark Rounsevell (United Kingdom/Germany), Almut Arneth (Germany), Katherine Calvin (The United States of America), Edouard Davin (France/Switzerland), Jan Fuglestvedt (Norway), Joanna House (United Kingdom), Alexander Popp (Germany), Joana Portugal Pereira (United Kingdom), Prajal Pradhan (Nepal/Germany), Jim Skea (United Kingdom), David Viner (United Kingdom).

'''About this box'''

The land-climate system is complex and future changes are uncertain, but methods exist (collectively known as futures analysis) to help decision-makers in navigating through this uncertainty. Futures analysis comprises a number of different and widely used methods, such as scenario analysis (Rounsevell and Metzger 2010 <sup>[[#fn:r444|444]]</sup> ), envisioning or target setting (Kok et al. 2018 <sup>[[#fn:r445|445]]</sup> ), pathways analysis (IPBES 2016 <sup>[[#fn:r446|446]]</sup> ; IPCC 2018 <sup>[[#fn:r447|447]]</sup> ) <sup>[[#fn:1|1]]</sup> , and conditional probabilistic futures (Vuuren et al. 2018 <sup>[[#fn:r448|448]]</sup> ; Engstrom et al. 2016 <sup>[[#fn:r449|449]]</sup> ; Henry et al. 2018 <sup>[[#fn:r450|450]]</sup> ) (Table 1 in this Cross-Chapter Box). Scenarios and other methods to characterise the future can support a discourse with decision-makers about the sustainable development options that are available to them. All chapters of this assessment draw conclusions from futures analysis and so, the purpose of this box is to outline the principal methods used, their application domains, their uncertainties and their limitations.

'''Exploratory scenario analysis'''

Many exploratory scenarios are reported in climate and land system studies on climate change (Dokken 2014 <sup>[[#fn:r451|451]]</sup> ), such as related to land-based, climate change mitigation via reforestation/afforestation, avoided deforestation or bioenergy (Kraxner et al. 2013 <sup>[[#fn:r452|452]]</sup> ; Humpenoder et al. 2014 <sup>[[#fn:r453|453]]</sup> ; Krause et al. 2017 <sup>[[#fn:r454|454]]</sup> ) and climate change impacts and adaptation (Warszawski et al. 2014 <sup>[[#fn:r455|455]]</sup> ). There are global-scale scenarios of food security (Foley et al. 2011 <sup>[[#fn:r456|456]]</sup> ; Pradhan et al. 2013 <sup>[[#fn:r457|457]]</sup> , 2014 <sup>[[#fn:r458|458]]</sup> ), but fewer scenarios of desertification, land degradation and restoration (Wolff et al. 2018 <sup>[[#fn:r459|459]]</sup> ). Exploratory scenarios combine qualitative ‘storylines’ or descriptive narratives of the underlying causes (or drivers) of change (Nakicenovic and Swart 2000 <sup>[[#fn:r460|460]]</sup> ; Rounsevell and Metzger 2010 <sup>[[#fn:r461|461]]</sup> ; O’Neill et al. 2014 <sup>[[#fn:r462|462]]</sup> ) with quantitative projections from computer models. Different types of models are used for this purpose based on very different modelling paradigms, baseline data and underlying assumptions (Alexander et al. 2016a <sup>[[#fn:r463|463]]</sup> ; Prestele et al. 2016 <sup>[[#fn:r464|464]]</sup> ). Figure 1 in this Cross-Chapter Box below outlines how a combination of models can quantify these components as well as the interactions between them.

Exploratory scenarios often show that socio-economic drivers have a larger effect on land-use change than climate drivers (Harrison et al. 2014 <sup>[[#fn:r465|465]]</sup> , 2016 <sup>[[#fn:r466|466]]</sup> ). Of these, technological development is critical in affecting the production potential (yields) of food and bioenergy and the feed conversion efficiency of livestock (Rounsevell et al. 2006 <sup>[[#fn:r467|467]]</sup> ; Wise et al. 2014 <sup>[[#fn:r|]]</sup> 468; Kreidenweis et al. 2018 <sup>[[#fn:r469|469]]</sup> ), as well as the area of land needed for food production (Foley et al. 2011 <sup>[[#fn:r470|470]]</sup> ; Weindl et al. 2017 <sup>[[#fn:r471|471]]</sup> ; Kreidenweis et al. 2018 <sup>[[#fn:r472|472]]</sup> ). Trends in consumption, for example, diets or waste reduction, are also fundamental in affecting land-use change (Pradhan et al. 2013 <sup>[[#fn:r473|473]]</sup> ; Alexander et al. 2016b <sup>[[#fn:r474|474]]</sup> ; Weindl et al. 2017 <sup>[[#fn:r475|475]]</sup> ; Alexander et al. 2017 <sup>[[#fn:r476|476]]</sup> ; Vuuren et al. 2018 <sup>[[#fn:r477|477]]</sup> ; Bajželj et al. 2014 <sup>[[#fn:r478|478]]</sup> ). Scenarios of land-based mitigation through large-scale bioenergy production and afforestation often lead to negative trade-offs with food security (food prices), water resources and biodiversity (Cross-Chapter Box 7 in Chapter 6).

Many exploratory scenarios are based on common frameworks such as the Shared Socio-economic Pathways (SSPs) (Popp et al. 2016 <sup>[[#fn:r479|479]]</sup> ; Riahi et al. 2017 <sup>[[#fn:r480|480]]</sup> ; Doelman et al. 2018 <sup>[[#fn:r481|481]]</sup> )) (Section 1.2). However, other methods are used. Stylised scenarios prescribe assumptions about climate and land-use change solutions, for example, dietary change, food waste reduction and afforestation areas (Pradhan et al. 2013 <sup>[[#fn:r482|482]]</sup> , 2014 <sup>[[#fn:r483|483]]</sup> ; Kreidenweis et al. 2016 <sup>[[#fn:r484|484]]</sup> ; Rogelj et al. 2018b <sup>[[#fn:r485|485]]</sup> ; Seneviratne et al. 2018 <sup>[[#fn:r486|486]]</sup> ; Vuuren et al. 2018 <sup>[[#fn:r487|487]]</sup> ). These scenarios provide useful thought experiments, but the feasibility of achieving the stylised assumptions is often unknown. Shock scenarios explore the consequences of low probability, high-impact events such as pandemic diseases, cyber-attacks and failures in food supply chains (Challinor et al. 2018 <sup>[[#fn:r488|488]]</sup> ), often in food security studies. Because of the diversity of exploratory scenarios, attempts have been made to categorise them into ‘archetypes’ based on the similarity between their assumptions in order to facilitate communication (IPBES 2018a <sup>[[#fn:r489|489]]</sup> ).

Conditional probabilistic futures explore the consequences of model parameter uncertainty in which these uncertainties are conditional on scenario assumptions (Neill 2004 <sup>[[#fn:r490|490]]</sup> ). Only a few studies have applied the conditional probabilistic approach to land-use futures (Brown et al. 2014 <sup>[[#fn:r491|491]]</sup> ; Engstrom et al. 2016 <sup>[[#fn:r492|492]]</sup> ; Henry et al. 2018 <sup>[[#fn:r493|493]]</sup> ). By accounting for uncertainties in key drivers these studies show large ranges in land-use change, for example, global cropland areas of 893–2380 Mha by the end of the 21st century (Engstrom et al. 2016 <sup>[[#fn:r494|494]]</sup> ). They also find that land-use targets may not be achieved, even across a wide range of scenario parameter settings, because of trade-offs arising from the competition for land (Henry et al. 2018 <sup>[[#fn:r495|495]]</sup> ; Heck et al. 2018 <sup>[[#fn:r496|496]]</sup> ). Accounting for uncertainties across scenario assumptions can lead to convergent outcomes for land-use change, which implies that certain outcomes are more robust across a wide range of uncertain scenario assumptions (Brown et al. 2014 <sup>[[#fn:r497|497]]</sup> ).

In addition to global scale scenario studies, sub-national studies demonstrate that regional climate change impacts on the land system are highly variable geographically because of differences in the spatial patterns of both climate and socio-economic change (Harrison et al. 2014 <sup>[[#fn:r498|498]]</sup> ). Moreover, the capacity to adapt to these impacts is strongly dependent on the regional, socio-economic context and coping capacity (Dunford et al. 2014 <sup>[[#fn:r499|499]]</sup> ); processes that are difficult to capture in global scale scenarios. Regional scenarios are often co-created with stakeholders through participatory approaches (Kok et al. 2014 <sup>[[#fn:r500|500]]</sup> ), which are powerful in reflecting diverse worldviews and stakeholder values. Stakeholder participatory methods provide additional richness and context to storylines, as well as providing salience and legitimacy for local stakeholders (Kok et al. 2014).

<div id="section-1-2-2-2-nature-and-scope-of-uncertainties-related-to-land-use-block-2"></div>

<span id="cross-chapter-box-1-table-1"></span>
====== Cross-Chapter Box 1, Table 1 ======

<span id="description-of-the-principal-methods-used-in-land-and-climate-futures-analysis."></span>
==== Description of the principal methods used in land and climate futures analysis. ====

{| class="wikitable"
|-

Futures method

Description and subtypes

Application domain

Time horizon

Examples in this assessment

|-

Exploratory scenarios.

Trajectories of change in system components from the present to contrasting, alterna- tive futures based on plausible and internally consistent assumptions about the underlying drivers of change

Long-term projections quantified with models

Climate system, land system and other components of the environment (e.g., biodiversity, ecosystem function- ing, water resources and quality), for example the SSPs

10–100 years

2.3, 2.6.2, 5.2.3, 6.1.4, 6.4.4, 7.2

|-

Business-as-usual scenarios

(including ‘outlooks’)

A continuation into the future of current trends<br />
in key drivers to explore the consequences of these in the near term

5–10 years, 20–30 years for outlooks

1.2.1, 2.6.2, 5.3.4, 6.1.4

|-

Policy and planning scenarios

(including business planning)

Ex ante analysis of the consequences of alternative policies or decisions based on known policy options or already implemented policy and planning measures

5–30 years

2.6.3, 5.5.2, 5.6.2, 6.4.4

|-

Stylised scenarios (with single and multiple options)

Afforestation/reforestation areas, bioenergy areas, protected areas for conservation, consumption patterns (e.g., diets, food waste)

10–100 years

2.6.1, 5.5.1, 5.5.2, 5.6.1, 5.6.2, 6.4.4, 7.2

|-

Shock scenarios (high impact single events)

Food supply chain collapses, cyberattacks, pandemic diseases (humans, crops and livestock)

Near-term events<br />
(up to 10 years) leading to long-term impacts (10–100 years)

5.8.1

|-

Conditional probabilistic futures

ascribe probabilities to uncertain drivers that are conditional on scenario assumptions

Where some knowledge is known about driver uncertainties, for example, population, economic growth, land-use change

10–100 years

1.2

|-

Normative scenarios.

Desired futures or outcomes that are aspirational and how to achieve them

Visions, goal-seeking or target-seeking scenarios

Environmental quality, societal development, human well-being, the Representative Concentration Pathways (RCPs,) 1.5°C scenarios

5–10 years to 10–100 years

2.6.2, 6.4.4, 7.2, 5.5.2

|-

Pathways as alternative sets<br />
of choices, actions or behaviours that lead to a future vision<br />
(goal or target)

Socio-economic systems, governance and policy actions

5–10 years to 10–100 years

5.5.2, 6.4.4, 7.2

|}

<div id="section-1-2-2-2-nature-and-scope-of-uncertainties-related-to-land-use-block-3"></div>

<span id="cross-chapter-box-figure-1"></span>
====== Cross-Chapter-Box-Figure-1 ======

<span id="interactions-between-land-and-climate-system-components-and-models-in-scenario-analysis.-the-blue-text-describes-selected-model-inputs-and-outputs."></span>
==== Interactions between land and climate system components and models in scenario analysis. The blue text describes selected model inputs and outputs. ====

[[File:6afc4f9f39c608c9b3bf3cbe99b59ea4 C1_Cross-Chapter-Box-Figure-1_Raw.jpg|thumb|400x300px]]

Interactions between land and climate system components and models in scenario analysis. The blue text describes selected model inputs and outputs.

<div id="section-1-2-2-2-nature-and-scope-of-uncertainties-related-to-land-use-block-4"></div>

Normative scenarios: visions and pathways analysis

Normative scenarios reflect a desired or target-seeking future. Pathways analysis is important in moving beyond the ‘what if?’ perspective of exploratory scenarios to evaluate how normative futures might be achieved in practice, recognising that multiple pathways may achieve the same future vision. Pathways analysis focuses on consumption and behavioural changes through transitions and transformative solutions (IPBES 2018a <sup>[[#fn:r501|501]]</sup> ). Pathways analysis is highly relevant in support of policy, since it outlines sets of time-dependent actions and decisions to achieve future targets, especially with respect to sustainable development goals, as well as highlighting trade-offs and co-benefits (IPBES 2018a <sup>[[#fn:r502|502]]</sup> ). Multiple, alternative pathways have been shown to exist that mitigate trade-offs whilst achieving the priorities for future sustainable development outlined by governments and societal actors. Of these alternatives, the most promising focus on long-term societal transformations through education, awareness raising, knowledge sharing and participatory decision-making (IPBES 2018a <sup>[[#fn:r503|503]]</sup> ).

What are the limitations of land-use scenarios?

Applying a common scenario framework (e.g., RCPs/SSPs) supports the comparison and integration of climate- and land-system scenarios, but a ‘climate-centric’ perspective can limit the capacity of these scenarios to account for a wider range of land-relevant drivers (Rosa et al. 2017 <sup>[[#fn:r504|504]]</sup> ). For example, in climate mitigation scenarios it is important to assess the impact of mitigation actions on the broader environment such as biodiversity, ecosystem functioning, air quality, food security, desertification/degradation and water cycles (Rosa et al. 2017 <sup>[[#fn:r505|505]]</sup> ). This implies the need for a more encompassing and flexible approach to creating scenarios that considers other environmental aspects, not only as a part of impact assessment, but also during the process of creating the scenarios themselves.

A limited number of models can quantify global scale, land-use change scenarios, and there is large variance in the outcomes of these models (Alexander et al. 2016a <sup>[[#fn:r506|506]]</sup> ; Prestele et al. 2016 <sup>[[#fn:r507|507]]</sup> ). In some cases, there is greater variability between the models themselves than between the scenarios that they are quantifying, and these differences vary geographically (Prestele et al. 2016 <sup>[[#fn:r508|508]]</sup> ). These differences arise from variations in baseline datasets, thematic classes and modelling paradigms (Alexander et al. 2016a <sup>[[#fn:r509|509]]</sup> ; Popp et al. 2016 <sup>[[#fn:r510|510]]</sup> ; Prestele et al. 2016 <sup>[[#fn:r511|511]]</sup> ). Model evaluation is critical in establishing confidence in the outcomes of modelled futures (Ahlstrom et al. 2012 <sup>[[#fn:r512|512]]</sup> ; Kelley et al. 2013 <sup>[[#fn:r513|513]]</sup> ). Some, but not all, land-use models are evaluated against observational data and model evaluation is rarely reported. Hence, there is a need for more transparency in land-use modelling, especially in evaluation and testing, as well as making model code available with complete sets of scenario outputs (e.g., Dietrich et al. 2018 <sup>[[#fn:r514|514]]</sup> ).

There is a small, but growing literature on quantitative pathways to achieve normative visions and their associated trade-offs (IPBES 2018a <sup>[[#fn:r515|515]]</sup> ). Whilst the visions themselves may be clearly articulated, the societal choices, behaviours and transitions needed to attain them, are not. Better accounting for human behaviour and decision-making processes in global scale land-use models would improve the capacity to quantify pathways to sustainable futures (Rounsevell et al. 2014 <sup>[[#fn:r516|516]]</sup> ; Arneth et al. 2014 <sup>[[#fn:r517|517]]</sup> ; Calvin and Bond-Lamberty 2018 <sup>[[#fn:r518|518]]</sup> ). It is, however, difficult to understand and represent human behaviour and social interaction processes at global scales. Decision-making in global models is commonly represented through economic processes (Arneth et al. 2014 <sup>[[#fn:r519|519]]</sup> ). Other important human processes for land systems including equity, fairness, land tenure and the role of institutions and governance, receive less attention, and this limits the use of global models to quantify transformative pathways, adaptation and mitigation (Arneth et al. 2014 <sup>[[#fn:r520|520]]</sup> ; Rounsevell et al. 2014 <sup>[[#fn:r521|521]]</sup> ; Wang et al. 2016 <sup>[[#fn:r|]]</sup> 522). No model exists at present to represent complex human behaviours at the global scale, although the need has been highlighted (Rounsevell et al. 2014 <sup>[[#fn:r523|523]]</sup> ; Arneth et al. 2014 <sup>[[#fn:r524|524]]</sup> ; Robinson et al. 2017 <sup>[[#fn:r525|525]]</sup> ; Brown et al. 2017 <sup>[[#fn:r526|526]]</sup> ; Calvin and Bond-Lamberty 2018 <sup>[[#fn:r527|527]]</sup> ).

<div id="section-1-2-2-3-uncertainties-in-decision-making"></div>

<span id="uncertainties-in-decision-making"></span>