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

== Box 4.3 Responses to Sea Level Rise ==

<div id="section-4-4-1introduction-block-1"></div>

'''''Protection''''' reduces coastal risk and impacts by blocking the inland propagation and other effects of mean or extreme sea levels (ESL). This includes: i) '''''hard protection''''' such as dikes, seawalls, breakwaters, barriers and barrages to protect against flooding, erosion and salt water intrusion (Nicholls, 2018), ii) '''''sediment-based protection''''' such as beach and shore nourishment, dunes (also referred to as soft structures), and iii) ecosystem-based adaptation (EbA) (see below). The three subcategories are often applied in combination as so-called hybrid measures. Examples are a marsh green-belt in front of a sea wall, or a sea wall especially designed to include niches for habitat formation (Coombes et al., 2015).

'''''Accommodation''''' includes diverse biophysical and institutional responses that mitigate coastal risk and impacts by reducing the vulnerability of coastal residents, human activities, ecosystems and the built environment, thus enabling the habitability of coastal zones despite increasing levels of hazard occurrence. Accommodation measures for erosion and flooding include building codes, raising house elevation (e.g., on stilts), lifting valuables to higher floors and floating houses and gardens (Trang, 2016). Accommodation measures for salinity intrusion include changes in land use (e.g., rice to brackish/salt shrimp aquaculture) or changes to salt tolerant crop varieties. Institutional accommodation responses include EWS, emergency planning, insurance schemes and setback zones (Nurse et al., 2014; Wong et al., 2014).

'''''Advance''''' creates new land by building seaward, reducing coastal risks for the hinterland and the newly elevated land. This includes land reclamation above sea levels by land filling with pumped sand or other fill material, planting vegetation with the specific intention to support natural accretion of land and surrounding low areas with dikes, termed polderisation, which also requires drainage and often pumping systems (Wang et al., 2014; Donchyts et al., 2016).

'''''Retreat''''' reduces coastal risk by moving exposed people, assets and human activities out of the coastal hazard zone. This includes the following three forms: i) '''''Migration,''''' which is the voluntary permanent or semi-permanent movement by a person at least for one year (Adger et al., 2014). ii) '''''Displacement,''''' which refers to the involuntary and unforeseen movement of people due to environment-related impacts or political or military unrest (Black et al., 2013; Islam and Khan, 2018; McLeman, 2018; Mortreux et al., 2018). iii) '''''Relocation''''' , also termed resettlement, managed retreat or managed realignment, which is typically initiated, supervised and implemented by governments from national to local levels and usually involves small sites and/or communities (Wong et al., 2014; Hino et al., 2017; Mortreux et al., 2018). Managed realignment may also be conducted for the purpose of creating new habitat. These three sub-categories are not neatly separable– any household’s decision to retreat may be ‘voluntary’ in theory, but in practice, may result from very limited choices. Displacement certainly occurs in response to extreme events but some of those retreating may have other options. Relocation programs may rely on incentives such as land buyouts that households adopt voluntarily. The need for retreat and other response measures can be reduced by avoiding new development commitments in areas prone to severe SLR hazards (Section 4.4.4.2)

'''''Ecosystem-based adaptation (EbA)''''' responses provide a combination of protect and advance benefits based on the sustainable management, conservation and restoration of ecosystems (Van Wesenbeeck et al., 2017). Examples include the conservation or restoration of coastal ecosystems such as wetlands and reefs. EbA measures protect the coastline by (i) attenuating waves, and, in the case of wetlands storm surge flows, by acting as obstacles and providing retention space (Krauss et al., 2009; Zhang et al., 2012; Vuik et al., 2015; Rupprecht et al., 2017); and (ii) by raising elevation and reducing rates of erosion through trapping and stabilising coastal sediments (Shepard et al., 2011), as well as building-up of organic matter and detritus (Shepard et al., 2011; McIvor et al., 2012a; McIvor et al., 2012b; Cheong et al., 2013; McIvor et al., 2013; Spalding et al., 2014). EbA is also referred to by various other names, including Natural and Nature-based Features, Nature-based Solutions, Ecological Engineering, Ecosystem-based Disaster Risk Reduction or Green Infrastructure (Bridges, 2015; Pontee et al., 2016).

<div id="section-4-4-1introduction-block-2"></div>

<span id="box-4.3-figure-1"></span>
<!-- START IMG -->
<!-- IMG TITLE -->
'''Box 4.3, Figure 1'''

<span id="box-4.3-figure-1-different-types-of-responses-to-coastal-risk-and-sea-level-rise-slr."></span>
<!-- IMG CAPTION -->
'''Box 4.3, Figure 1 | Different types of responses to coastal risk and sea level rise (SLR).'''
<!-- IMG FILE -->
[[File:72ef5e64e9bbe7f62946c9abe225fe7a IPCC-SROCC-CH_4_Box_4_3_figure_1-3000x1124.jpg]]

Box 4.3, Figure 1 | Different types of responses to coastal risk and sea level rise (SLR).

<!-- END IMG -->
<span id="observed-and-projected-responses-their-costs-benefits-co-benefits-drawbacks-efficiency-and-governance"></span>
=== 4.4.2 Observed and Projected Responses, their Costs, Benefits, Co-benefits, Drawbacks, Efficiency and Governance ===

<div id="section-4-4-2-1types-of-responses-and-framework-for-assessment"></div>

<span id="types-of-responses-and-framework-for-assessment"></span>
==== 4.4.2.1 Types of Responses and Framework for Assessment ====

<div id="section-4-4-2-1types-of-responses-and-framework-for-assessment-block-1"></div>

Following earlier IPCC Reports Protection, Retreat and Accommodation responses to SLR and its impacts are distinguished between (Nicholls et al., 2007; Wong et al., 2014), and Advance is added as a fourth type of response that consists in building seaward and upward (Box 4.3). Advance had not received much attention in the climate change literature but plays an important role in coastal development across the world (e.g., Institution of Civil Engineers, 2010; Lee, 2014; Donchyts et al., 2016). The broader term response is used here instead of adaptation, because some responses such as retreat may or may not be meaningfully considered to be adaptation (Hinkel et al., 2018). Responses that address the causes of climate change, such as mitigating GHGs or geoengineering temperature and sea level responses to emissions fall beyond the scope of this chapter, and are addressed in SR1.5 (Hoegh-Guldberg et al., 2018). In coastal areas where anthropogenic subsidence contributes to relative SLR, another important type of response is the management of subsidence by, for instance, restricting ground fluid abstraction. Although this type of measure is considered in the risk assessment developed in Section 4.3.4, it is not assessed here due to a lack of space.

Observed coastal responses are rarely responses to climate-change induced SLR only, but also to relative SLR caused by land subsidence as well as current coastal risks and many socioeconomic factors and related hazards. As a consequence, coastal responses have been practised for centuries, and there are many experiences specifically in places that have subsided up to several metres due to earthquakes or anthropogenic ground fluid abstraction in the last century that responding to climate-change induced SLR can draw upon (Esteban et al., 2019). Finally, in practise, many responses are hybrid, applying combinations of protection, accommodation, retreat, advance and EbA.

Since AR5, the literature on SLR responses has grown significantly. It is assessed in this section for the five above-described broad types of responses in terms of the following six criteria:

* '''''Observed responses''''' across geographies, describing where the different types of responses have been implemented.
* '''''Projected responses''''' , which refers to the potential extent of responses in the future, as assessed in the literature through modelling or in a more qualitative way.
* '''''Cost of responses''''' , which refers to the costs of implementing and maintaining responses. Other costs that arise due to negative side-effects of implementing a response are captured under the criterion ‘co-benefits and drawbacks’.
* '''''Effectiveness of responses''''' in terms of reducing SLR risks and impacts. This includes biophysical and technical limits beyond which responses cease to be effective.
* '''''Co-benefits and drawbacks''''' of responses that occur next to the intended benefits of reducing SLR risks and impacts.
* '''''Governance challenges (or barriers)''''' , which refers to institutional and organisational factors that have been found to hinder the effective, efficient and equitable implementation of responses (see also Section 4.4.3).
* '''''Economic efficiency''''' of responses, which refers to the overall monetised balance of costs, benefits (in terms of the effectiveness of responses), co-benefits and drawbacks. Economic barriers arise if responses have a negative net benefit or a benefit-cost ratio smaller than one. While it would be desirable to have information on the economic efficiency of integrated responses combining different response types, an assessment cannot be provided here due to the lack of literature.

<div id="section-4-4-2-2hard-and-sediment-based-protection"></div>

<span id="hard-and-sediment-based-protection"></span>
==== 4.4.2.2 Hard and Sediment-Based Protection ====

<div id="section-4-4-2-2hard-and-sediment-based-protection-block-1"></div>

<span id="observed-hard-and-sediment-based-protection-across-geographies"></span>
===== 4.4.2.2.1 Observed hard and sediment-based protection across geographies =====

Coastal protection through hard measures is widespread around the world, although it is difficult to provide estimates on how many people benefit from them. Currently, at least 20 million people living below normal high tides are protected by hard structures (and drainage) in countries such as Belgium, Canada, China, Germany, Italy, Japan, the Netherlands, Poland, Thailand, the UK, and the USA (Nicholls, 2010). Many more people living above high tides are also protected against ESL by hard structures in major cities around the world. There is a concentration of these measures in northwest Europe and East Asia, although extensive defences are also found in and around many coastal cities and deltas. For example, large scale coastal protection exists in Vancouver (Canada), Alexandria (Egypt) and Keta (Ghana; Nairn et al., 1999 <sup>[[#fn:r1544|1544]]</sup> ) and 6000 km of polder dikes in coastal Bangladesh. Gittman et al. (2015) estimate that 14% of the total US coastline has been armoured, with New Orleans being an example of an area below sea level dependent on extensive engineered protection (Kates et al., 2006 <sup>[[#fn:r1545|1545]]</sup> ; Rosenzweig and Solecki, 2014 <sup>[[#fn:r1546|1546]]</sup> ; Cooper et al., 2016 <sup>[[#fn:r1547|1547]]</sup> ). Defences built and raised for tsunami protection, such as post-2011 in Japan (Raby et al., 2015 <sup>[[#fn:r1548|1548]]</sup> ), also provide protection against SLR.

The application of sediment-based protection measures also has a long history, offering multiple benefits in terms of enhancing safety, recreation and natural systems (JSCE, 2000 <sup>[[#fn:r1549|1549]]</sup> ; Dean, 2002 <sup>[[#fn:r1550|1550]]</sup> ; Hanson et al., 2002 <sup>[[#fn:r1551|1551]]</sup> ; Cooke et al., 2012 <sup>[[#fn:r1552|1552]]</sup> ). About 24% of the world’s sandy beaches are currently eroding by rates faster than 0.5 m yr–1 (Luijendijk et al., 2018 <sup>[[#fn:r1553|1553]]</sup> ). In the USA, Europe and Australia, these responses are often driven by the recreational value of beaches and the high economic benefits associated with beach tourism. More recently, sediment-based measures are implemented as effective and yet flexible measures to address SLR (Kabat et al., 2009 <sup>[[#fn:r1554|1554]]</sup> ) and experiments are being conducted with innovative decadal scale application of sediments such as the sand engine in the Netherlands (Stive et al., 2013 <sup>[[#fn:r1555|1555]]</sup> ).

There is high confidence that most major upgrades in defences happen after coastal disasters (Box 4.1). Dikes were raised and reienforced after the devastating coastal flood of 1953 in the Netherlands and the UK, and in 1962 in Germany. In New Orleans, investments in the order of 15 billion USD, including a major storm surge barrier, followed Hurricane Katrina in 2005 (Fischetti, 2015 <sup>[[#fn:r1556|1556]]</sup> ), and in New York the Federal Government made available 16 billion USD for disaster recovery and adaptation after Superstorm Sandy in 2012 (NYC, 2015). Examples in which SLR has been considered proactively in the planning process include SLR safety margins in, for example, the UK, Germany and France, upgrading defences according to cost-benefit analysis in the Netherlands, and SLR guidance in the USA (USACE, 2011 <sup>[[#fn:r1557|1557]]</sup> ).

<div id="section-4-4-2-2hard-and-sediment-based-protection-block-2"></div>

<span id="projected-hard-and-sediment-based-protection"></span>
===== 4.4.2.2.2 Projected hard and sediment-based protection =====

There is ''high confidence'' that hard coastal protection will continue to be a widespread response to SLR in densely populated and urban areas during the 21st century, because this response is widely practised (Section 4.4.2.2.2), effective in reducing current (Section 4.4.2.2.2) and future flood risk (Section 4.3.3.2) and highly cost efficient in urban and densely populated areas (Section 4.4.2.7). There is, however, ''low agreement'' on the level of hard coastal protections to expect, with projections being based on different assumptions. A model assuming that coastal societies upgrade hard protection following scenario-based cost-benefit analysis finds that 22% of the global coastline will be protected under various SSPs and 1 m of 21st century global mean SLR (Nicholls et al., 2019 <sup>[[#fn:r1558|1558]]</sup> ). Another model assuming that only areas for which benefit-cost ratios are above 1 under SLR scenarios up to 2 m, all SSPs and discount rates up to 6%, finds that this would lead to protecting 13% of the global coastline (Lincke and Hinkel, 2018 <sup>[[#fn:r1559|1559]]</sup> ; Figure 4.14).

<div id="section-4-4-2-2hard-and-sediment-based-protection-block-3"></div>

<span id="cost-of-hard-and-sediment-based-protection"></span>
===== 4.4.2.2.3 Cost of hard and sediment-based protection =====

There is ''medium evidence'' and ''medium agreement'' on the costs of hard protection. Data on the costs of hard defences is only available for few countries and unit costs estimated from this data vary substantially depending on building/fill material used, labour cost, urban versus rural settings, hydraulic loads, etc. (Jonkman et al., 2013 <sup>[[#fn:r1587|1587]]</sup> ; Lenk et al., 2017 <sup>[[#fn:r1588|1588]]</sup> ; Aerts, 2018 <sup>[[#fn:r1589|1589]]</sup> ; Nicholls et al., 2019 <sup>[[#fn:r1590|1590]]</sup> ). In general, there has been limited systematic data collection across sites, although useful national guidance does exist in some cases (Environment Agency, 2015 <sup>[[#fn:r1591|1591]]</sup> ). Defences depend on good maintenance to remain effective. For some types of infrastructure such as surge barriers, maintenance costs are poorly described and hence more uncertain (Nicholls et al., 2007 <sup>[[#fn:r1592|1592]]</sup> ). Protection-based adaptation to saltwater intrusion is more complex than adaptation to flooding and erosion, and there is less experience to draw upon.

Based on these unit cost estimates, and different assumptions on future protection, global annual protection costs have been estimated to be 12–71 billion USD considering coastal dikes only (Hinkel et al., 2014 <sup>[[#fn:r1593|1593]]</sup> ) and about 40–170 billion USD yr -1 considering coastal dikes, river dikes and storm surge barriers, under RCP2.6, and about 25–200 billion USD yr -1 considering coastal dikes only (Tamura et al. 2019 <sup>[[#fn:r1594|1594]]</sup> ) under RCP8.5. If protection is widely practised through the 21st century, the bulk of the costs will be maintenance rather than capital costs (Nicholls et al., 2019 <sup>[[#fn:r1595|1595]]</sup> ).

<span id="table-4.7"></span>
<!-- START TABLE -->
'''Table 4.7'''

'''Table 4.7:''' Capital and maintenance costs of hard protection measures.

<!-- TABLE -->
{| class="wikitable"
|-
| Measure
| Capital cost (in million USD unless stated otherwise)
| Annual Maintenance Cost (% of capital cost)
|-
| Sea Wall
| 0.4–27.5 km -1 length and metre height (Linham et al., 2010)
| 1–2% per annum (Jonkman et al., 2013)
|-
| Sea Dike
| 0.9–69.9 km -1 length and metre height (Jonkman et al., 2013; Nicholls et al., 2019; Tamura et al., 2019)
| 1–2% per annum (Jonkman et al., 2013)
|-
| Breakwater
| 2.5–10.0 km -1 length (Narayan et al., 2016)
| 1% per annum (Jonkman et al., 2013)
|-
| Storm Surge Barrier
| 0.9–2.7 (Jonkman et al., 2013) or 2.2 (Mooyaart and Jonkman, 2017) million EUR per metre width
| 1% per annum (Mooyaart and Jonkman, 2017) or 5–10% per annum (Nicholls et al., 2007)
|-
| Saltwater
Intrusion Barriers

| Limited knowledge
|}
<!-- END TABLE -->

<div id="section-4-4-2-2hard-and-sediment-based-protection-block-4"></div>

Sediment-based measures are generally costed as the unit cost of sand (or gravel) delivery multiplied by the volumetric demand. Unit costs range from 3–21 USD m <sup>–</sup> ³ sand , with some high outlier costs in, for example, the UK, South Africa and New Zealand (Linham et al., 2010 <sup>[[#fn:r1596|1596]]</sup> ; Aerts, 2018 <sup>[[#fn:r1597|1597]]</sup> ). Costs are small where sources of sand are plentiful and close to the sites of demand. Costs are further reduced by shoreface nourishment approaches. The Netherlands maintains its entire open coast with large-scale shore nourishment (Mulder et al., 2011 <sup>[[#fn:r1598|1598]]</sup> ) and the innovative sand engine has been implemented as a full-scale decadal experiment (Stive et al., 2013 <sup>[[#fn:r1599|1599]]</sup> ). The capital costs for dunes are similar to beach nourishment, although placement and planting vegetation may raise costs. Maintenance costs vary from almost nothing to several million USD km <sup>–1</sup> , although costs are usually at the lower end of this range (Environment Agency, 2015 <sup>[[#fn:r1600|1600]]</sup> ).

<div id="section-4-4-2-2hard-and-sediment-based-protection-block-5"></div>

<span id="effectiveness-of-hard-and-sediment-based-protection"></span>
===== 4.4.2.2.4 Effectiveness of hard and sediment-based protection =====

There is ''high confidence'' that well designed and maintained hard and sediment-based protection is very effective in reducing risk to the impacts of SLR and ESL (Horikawa, 1978 <sup>[[#fn:r1572|1572]]</sup> ; USACE, 2002 <sup>[[#fn:r1673|1673]]</sup> ; CIRIA, 2007 <sup>[[#fn:r1674|1674]]</sup> ). This includes situations in which coastal megacities in river deltas have experienced, and adapted to, relative SLR of several metres caused by land subsidence during the 20th century (Kaneko and Toyota, 2011 <sup>[[#fn:r1675|1675]]</sup> ; Esteban et al., 2019 <sup>[[#fn:r1676|1676]]</sup> ; Box 4.1). In principle, there are no technological limits to protect the coast during the 21st century even under high-end SLR of 2 m (Section 4.3.3.2), but technological challenges can make protection very expensive and hence unaffordable in some areas (Hinkel et al., 2018 <sup>[[#fn:r1677|1677]]</sup> ). Examples include southeast Florida, because protected areas can be flooded by rising groundwater through underlying porous limestone (Bloetscher et al., 2011 <sup>[[#fn:r1678|1678]]</sup> ). Gradually rising water tables behind defences is also an issue, which can be managed by increasing pumping and drainage (Aerts, 2018 <sup>[[#fn:r1679|1679]]</sup> ). Maintaining this effectiveness over time requires regular monitoring and maintenance, accounting for changing conditions such as SLR and widespread erosional trends in front of the defences. There will always be residual risks, which can be reduced, but never eliminated, by engineering protection infrastructure to very high standards, such as so-called ‘unbreakable dikes’ (de Bruijn et al., 2013).

It is difficult to assess at what point in time and for which amount of SLR technical limits for coastal protection will be reached. Parts of Tokyo have been protected against five metres of relative SLR during the 21st century (Kaneko and Toyota, 2011 <sup>[[#fn:r1680|1680]]</sup> ) and it has been argued that it is possible to preserve territorial integrity of the Netherlands even under 5 m SLR, using current engineering technology (Aerts et al., 2008 <sup>[[#fn:r1681|1681]]</sup> ; Olsthoorn et al., 2008 <sup>[[#fn:r1682|1682]]</sup> ). This suggests that under RCP2.6, technical limits to adaptation will be rare even under longer-term SLR. Protecting against high-end SLR will be increasingly technically challenging as we move beyond the 21st century. This is not only due to the absolute amount of SLR, but also due to the very high rates of annual SLR (e.g., 10–20 mm yr –1 ''likely'' range under RCP8.5 in 2100), which challenge the planning and implementation of hard protection because major protection infrastructure requires decades to plan and implement (Gilbert et al., 1984 <sup>[[#fn:r1683|1683]]</sup> ; Burcharth et al., 2014 <sup>[[#fn:r1684|1684]]</sup> ). In summary, the higher and faster SLR, the more challenging coastal protection will be, but quantifying this is difficult. In any case, before technical limits are reached, economic and social limits will be reached because societies are neither economically able nor socially willing to invest in coastal protection (Sections 4.4.2.2 and 4.3.3.2; Hinkel et al., 2018; Esteban et al., 2019).

<div id="section-4-4-2-2hard-and-sediment-based-protection-block-6"></div>

<span id="co-benefits-and-drawbacks-of-hard-and-sediment-based-protection"></span>
===== 4.4.2.2.5 Co-benefits and drawbacks of hard and sediment-based protection =====

When space is limited (e.g., in an urban setting), co-benefits can be generated through multi-functional hard flood defences, which combine flood protection with other urban functions, such as car parks, buildings, roads or recreational spaces into one multifunctional structure (Stalenberg, 2013 <sup>[[#fn:r1601|1601]]</sup> ; van Loon-Steensma and Vellinga, 2014 <sup>[[#fn:r162|162]]</sup> ). An important co-benefit of sediment-based protection, such as beach nourishment and dune management, is that it preserves beach and associated environments, as well as tourism (Everard et al., 2010 <sup>[[#fn:r1603|1603]]</sup> ; Hinkel et al., 2013a <sup>[[#fn:r1604|1604]]</sup> ; Stive et al., 2013 <sup>[[#fn:r1605|1605]]</sup> ).

Drawbacks of hard protection include the alteration of hydrodynamic and morphodynamic patterns, which in turn may export flooding and erosion problems downdrift (Masselink and Gehrels, 2015 <sup>[[#fn:r1606|1606]]</sup> ; Nicholls et al., 2015 <sup>[[#fn:r1607|1607]]</sup> ). For example, protection of existing shoreline in estuaries and tidal creeks may increase tidal amplification in the upper parts (Lee et al., 2017 <sup>[[#fn:r1608|1608]]</sup> ). Hard protection also hinders or prohibits the onshore migration of geomorphic features and ecosystems (called coastal squeeze; Pontee, 2013 <sup>[[#fn:r1609|1609]]</sup> ; Gittman et al., 2016 <sup>[[#fn:r1610|1610]]</sup> ), leading to both a loss of habitat as well as of the protection function of ecosystems (see Sections 4.3.2.4 and 4.4.2.2). Another drawback of raising hard structures, also emphasised in AR5, is the risk of lock-in to a development pathway in which development intensifies behind higher and higher defences, with escalating severe consequences in the event of protection failure (Wong et al., 2014 <sup>[[#fn:r1611|1611]]</sup> ; Welch et al., 2017 <sup>[[#fn:r1612|1612]]</sup> ), as experienced in Hurricane Katrina impacted New Orleans (Burby, 2006 <sup>[[#fn:r1613|1613]]</sup> ; Freudenburg et al., 2009 <sup>[[#fn:r1614|1614]]</sup> ). This lock-in results from protection attracting further economic development in the flood zone within defenses, which then leads to further raising defences with SLR, and the growing value of exposed assets.

Seabed dredging of sand and gravel can have negative impacts on marine ecosystems such as seagrass meadows and corals (Erftemeijer and Lewis III, 2006 <sup>[[#fn:r1615|1615]]</sup> ; Erftemeijer et al., 2012 <sup>[[#fn:r1616|1616]]</sup> ). Nourishment practices on sandy beaches have also been shown to have drawbacks for local ecosystems if local habitat factors are not taken into consideration when planning and implementing nourishment and maintenance (Speybroeck et al., 2006 <sup>[[#fn:r1617|1617]]</sup> ). A further emerging issue is beach material scarcity mainly driven by demand of sand and gravel for construction, but also for beach and shore nourishment (Peduzzi, 2014 <sup>[[#fn:r1618|1618]]</sup> ; Torres et al., 2017 <sup>[[#fn:r1619|1619]]</sup> ), which makes sourcing the increasing volumes of beach materials required to sustain beaches in the face of SLR more expensive and challenging (Roelvink, 2015 <sup>[[#fn:r1620|1620]]</sup> ).

<div id="section-4-4-2-2hard-and-sediment-based-protection-block-7"></div>

<span id="governance-of-hard-and-sediment-based-protection"></span>
===== 4.4.2.2.6 Governance of hard and sediment-based protection =====

Reviews and comparative case studies confirm findings of AR5 that governance challenges are amongst the most common hindrance to implementing coastal measures (Ekstrom and Moser, 2014 <sup>[[#fn:r1621|1621]]</sup> ; Hinkel et al., 2018 <sup>[[#fn:r1622|1622]]</sup> ). One main issue to resolve is conflicting stakeholder interests. This includes conflicts between those favouring protection and those being negatively affected by adaptation measures. In Catalonia, for example, the tourism sector welcomes beach nourishment because it provides direct benefits, whereas those dependent upon natural resources (e.g., fishermen) are increasingly in opposition because they fear that sand mining destroys coastal habitat and livelihood prospects (González-Correa et al., 2008 <sup>[[#fn:r1623|1623]]</sup> ).

There is also conflict related to the distribution of public money between communities receiving public support for adaptation and non-coastal communities who pay for this support through taxes (Elrick-Barr et al., 2015 <sup>[[#fn:r1624|1624]]</sup> ). Generally, access to financial resources for adaptation, including from public sources, development and climate finance or capital markets, frequently constrain adaptation (Ekstrom and Moser, 2014 <sup>[[#fn:r1625|1625]]</sup> ; Hinkel et al., 2018 <sup>[[#fn:r1626|1626]]</sup> ). For example, homeowners are often not willing to pay taxes or levies for public protection or sediment-base measures even if they directly benefit, as found, for example in communities on the US east coast where beach nourishment is used to maintain recreational and tourism amenities (Mullin et al., 2019 <sup>[[#fn:r1627|1627]]</sup> ). In many parts of the world, coastal adaptation governance is further complicated by existing conflicts over resources. For example, illegal coastal sand mining is currently a major driver of coastal erosion in many parts of the developing world (Peduzzi, 2014 <sup>[[#fn:r1628|1628]]</sup> ). Examples of this can be found in Ghana (Addo, 2015 <sup>[[#fn:r1629|1629]]</sup> ) and the Comoros (Betzold and Mohamed, 2017 <sup>[[#fn:r1630|1630]]</sup> ).

An associated governance challenge is ensuring the effective maintenance of coastal protection. Ineffective maintenance has contributed to many coastal disasters in the past, such as in New Orleans (Andersen, 2007 <sup>[[#fn:r1631|1631]]</sup> ). AR5 highlighted that effective maintenance is challenging in a small island context due to a lack of adequate funds, policies and technical skills (Nurse et al., 2014 <sup>[[#fn:r1632|1632]]</sup> ). In some countries in which coastal defence systems have a long history, effective governance arrangements for maintenance, such as the Water Boards in the Netherlands, have emerged. In Bangladesh, where Dutch-like polders were introduced in the 1960s, maintenance has been a challenge due to shifts in multi-level governance structures associated with independence, national policy priorities and donor involvement (Dewan et al., 2015 <sup>[[#fn:r1633|1633]]</sup> ).

<div id="section-4-4-2-2hard-and-sediment-based-protection-block-8"></div>

<span id="economics-of-coastal-adaptation"></span>
===== 4.4.2.2.7 Economics of coastal adaptation =====

At global scales, new economic assessments of responses have mostly focused on the direct costs of hard protection and the benefits of reducing coastal extreme event flood risks. These studies confirm AR5 findings that the benefits of reducing coastal flood risk through hard protection exceed the costs of protection, on a global average, and for cities and densely populated areas, during the 21st century even under high-end SLR ( ''medium evidence, high agreement'' ; Hallegatte et al., 2013 <sup>[[#fn:r1634|1634]]</sup> ; Wong et al., 2014 <sup>[[#fn:r1635|1635]]</sup> ; Diaz, 2016 <sup>[[#fn:r1636|1636]]</sup> ; Lincke and Hinkel, 2018 <sup>[[#fn:r1637|1637]]</sup> ). For example, Lincke and Hinkel (2018) find that, during the 21st century, it is economically efficient to protect 13% of the global coastline, which corresponds to 90% of global floodplain population, under SLR scenarios from 0.3–2.0 m, five SSPs and discount rates up to 6% (Figure 4.14). While the above two studies have not considered the effects of hard protection in reducing the area of coastal wetlands, it is expected that coastal hard protection in densely populated areas and conserving wetlands in sparsely populated areas can go hand in hand. Protecting less than 42% of the global coastline would leave coastal wetlands sufficient accommodation space to even grow in areas under rising sea levels during the 21st century (Schuerch et al., 2018 <sup>[[#fn:r1639|1639]]</sup> ). Diaz (2016), who includes the cost of wetland loss, using a simpler wetland model, finds that both protection and retreat reduce the global net present costs of SLR by a factor of seven as compared to no adaptation (applying a discount rate of 4%) under 21st century SLR of 0.3–1.3 m and SSP2. There is no global study that has considered social costs and benefits of responses (e.g., health, beach amenity, etc.) or looked at the economics of accommodate, retreat and advance responses.

<span id="figure-4.14"></span>
<!-- START IMG -->
<!-- IMG TITLE -->
'''Figure 4.14'''

<span id="figure-4.14-economic-robustness-of-coastal-protection-under-sea-level-rise-slr-scenarios-from-0.32.0-m-the-five-shared-socioeconomic-pathways-ssps-and-discount-rates-of-up-to-6.-coastlines-are-coloured-according-to-the-percentage-of-scenarios-under-which-the-benefit-cost-ratio-of-protection-reduced-flood-risk-divided-by-the-cost-of-protection"></span>
<!-- IMG CAPTION -->
'''Figure 4.14 | Economic robustness of coastal protection under sea level rise (SLR) scenarios from 0.3–2.0 m, the five Shared Socioeconomic Pathways (SSPs) and discount rates of up to 6%. Coastlines are coloured according to the percentage of scenarios under which the benefit-cost ratio of protection (reduced flood risk divided by the cost of protection) […]'''
<!-- IMG FILE -->
[[File:b867d55a5f3933dfe77ecb75a01cc583 IPCC-SROCC-CH_4_14-3000x1772.jpg]]

Figure 4.14 | Economic robustness of coastal protection under sea level rise (SLR) scenarios from 0.3–2.0 m, the five Shared Socioeconomic Pathways (SSPs) and discount rates of up to 6%. Coastlines are coloured according to the percentage of scenarios under which the benefit-cost ratio of protection (reduced flood risk divided by the cost of protection) are above 1. Source: Lincke and Hinkel (2018).

At local scales, a large number of economic assessments of response options are available but mostly in the grey literature and again with a focus on hard and sediment-based protection. Similar to the global studies, hard protection is generally found to be economically efficient for urban and densely populated areas such as New York, USA (Aerts et al., 2014 <sup>[[#fn:r1640|1640]]</sup> ) and Ho Chi Minh City, Vietnam (Scussolini et al., 2017 <sup>[[#fn:r1641|1641]]</sup> ). Both global and local studies show that sediment-based protection, such as beach nourishment is economically efficient in areas of intensive tourism development due to the large revenues generated within this sector (Rigall-I-Torrent et al., 2011 <sup>[[#fn:r1642|1642]]</sup> ; Hinkel et al., 2013a <sup>[[#fn:r1643|1643]]</sup> ).

<!-- END IMG -->
<div id="section-4-4-2-3ecosystem-based-adaptation"></div>

<span id="ecosystem-based-adaptation"></span>
==== 4.4.2.3 Ecosystem-based Adaptation ====

<div id="section-4-4-2-3ecosystem-based-adaptation-block-1"></div>

<span id="observed-ecosystem-based-adaptation-across-geographies"></span>
===== 4.4.2.3.1 Observed ecosystem-based adaptation across geographies =====

Relative to hard adaptation measures whose global distribution is not known in detail (Scussolini et al., 2015 <sup>[[#fn:r1644|1644]]</sup> ), the current global distribution of coastal ecosystems is well-studied (e.g., for saltmarshes and mangroves, respectively; Giri et al., 2011 <sup>[[#fn:r1645|1645]]</sup> ; Mcowen et al., 2017 <sup>[[#fn:r1646|1646]]</sup> ). EbA, by definition, can only exist and function where the environmental conditions are appropriate for a given ecosystem. Mangroves, salt marshes and reefs occur along about 40–50% of the world’s coastlines (Wessel and Smith, 1996 <sup>[[#fn:r1647|1647]]</sup> ; Burke, 2011 <sup>[[#fn:r1648|1648]]</sup> ; Giri et al., 2011 <sup>[[#fn:r1649|1649]]</sup> ; Mcowen et al., 2017 <sup>[[#fn:r1650|1650]]</sup> ). However, there is no clear estimate on the global length of coastline covered by ecosystems relevant for EbA in the face of SLR in part because of a mismatch between the spatial resolutions of different estimates available. Mangroves occur on tropical and subtropical coasts, and cover 138,000–152,000 km <sup>2</sup> across about 120 countries (Spalding et al., 2010 <sup>[[#fn:r1651|1651]]</sup> ; Giri et al., 2011 <sup>[[#fn:r1652|1652]]</sup> ). At least 150,000 km of coastline in over 100 countries benefit from the presence of coral reefs (Burke, 2011 <sup>[[#fn:r1653|1653]]</sup> ) and these are estimated to protect over 100 million people from wave-induced flooding globally (Ferrario et al., 2014 <sup>[[#fn:r1654|1654]]</sup> ). The extent of other coastal habitats is less well known: salt marshes are estimated to occur in 99 countries, especially in temperature to high latitude locations, with nearly 5,500,000 ha mapped across 43 countries (Mcowen et al., 2017 <sup>[[#fn:r1655|1655]]</sup> ).

Since AR5 there has been growing recognition of the value of conserving existing coastal ecosystems, and where possible restoring them, for the flood protection and multiple other benefits they provide (Temmerman et al., 2013 <sup>[[#fn:r1656|1656]]</sup> ; Arkema et al., 2015 <sup>[[#fn:r1657|1657]]</sup> ). In parallel, EbA measures are increasingly being incorporated and required within national plans, strategies and targets (Lo, 2016), international adaptation funding mechanisms, such as the Adaptation Fund (AF; e.g., in Sri Lanka and India; Epple et al., 2016 <sup>[[#fn:r1658|1658]]</sup> ), and national natural capital valuations (Beck and Lange, 2016 <sup>[[#fn:r1659|1659]]</sup> ). Given their relative novelty, there is widespread interest in building and collecting knowledge of EbA implementation case-studies and examples (Table 4.7). Meanwhile, coastal communities around the globe are already implementing EbA responses at local scales, with emphasis on community participation and ownership and local priorities, needs and capacities (Reid, 2016 <sup>[[#fn:r1660|1660]]</sup> ; see Section 4.4.4.4).

EbA has been used as an integral part of some retreat, advance and accommodation responses. For example, on coastlines where high-risk properties are relocated inland, space can be made for ecosystem restoration to enhance natural biodiversity and provide coastal protection (French, 2006 <sup>[[#fn:r1661|1661]]</sup> ; Coastal Protection and Restoration Authority of Louisiana, 2017 <sup>[[#fn:r1662|1662]]</sup> ). There are also examples of ecosystem restoration to advance coastlines and build land elevation (Chung, 2006 <sup>[[#fn:r1663|1663]]</sup> ). EbA can also be an element of accommodation responses by, for example, restoring or creating marshes to provide space for flood water (Temmerman et al., 2013 <sup>[[#fn:r1664|1664]]</sup> ).

<div id="section-4-4-2-3ecosystem-based-adaptation-block-2"></div>

<span id="projected-ecosystem-based-adaptation"></span>
===== 4.4.2.3.2 Projected ecosystem-based adaptation =====

While there are projections available of ecosystem responses to climate change and SLR (Section 4.3.3), to date, there are no large-scale projections available on the future extent of EbA. However, several coastal nations, particularly Small Island Developing States (SIDS) explicitly advocate EbA measures as a means to address future coastal hazard and SLR concerns. Based on Nationally Determined Contributions (NDCs) submitted to the United Nations Framework Convention on Climate Change (UNFCCC), more than 30 SIDS cite EbA as a preferred SLR response, with mangrove planting being the most common measure (Wong, 2018 <sup>[[#fn:r1665|1665]]</sup> ).

<div id="section-4-4-2-3ecosystem-based-adaptation-block-3"></div>

<span id="cost-of-ecosystem-based-adaptation"></span>
===== 4.4.2.3.3 Cost of ecosystem-based adaptation =====

There is ''limited evidence and low agreement'' on the costs of ecosystem-based measures to make generally valid estimations of the unit costs across large spatial scales. The total cost of an ecosystem-based measure includes capital costs, maintenance costs, the cost of land and, in some situations, permitting costs (Bilkovic, 2017 <sup>[[#fn:r1666|1666]]</sup> ). The costs of restoring and maintaining coastal habitats depend on coastal setting, habitat type and project conditions. In general, unit restoration costs are lowest for mangroves, higher for salt marshes and oyster reefs, and highest for seagrass beds and coral reefs (Table 4.8).

The conservation of coral reefs and other coastal habitats may also entail substantial opportunity costs because alternative uses of this land, such as through agricultural production, industry and settlements, are generally of high economic value (Stewart et al., 2003 <sup>[[#fn:r1667|1667]]</sup> ; Balmford et al., 2004 <sup>[[#fn:r1668|1668]]</sup> ; Adams et al., 2011 <sup>[[#fn:r1669|1669]]</sup> ; Hunt, 2013 <sup>[[#fn:r1670|1670]]</sup> ). The high value of these alternative uses are the reason why globally, coastal ecosystems are amongst the ecosystems that face the highest rates of anthropogenic destruction, with estimated annual losses of 1–3% of mangroves area, 2–5% seagrass area and 4–9% corals (Duarte et al., 2013 <sup>[[#fn:r1671|1671]]</sup> ). Conserving these areas means reversing these trends.

Under the right conditions, and to some extent, EbA measures are free of maintenance costs, because they respond and adapt to changes in their coastal environment. However, maintenance can become important in the aftermath of damage by storms or human action, for example, when wetlands and reefs can be damaged by high winds, waves and surges, or affected by dredging operations (Smith III et al., 2009 <sup>[[#fn:r1672|1672]]</sup> ; Puotinen et al., 2016 <sup>[[#fn:r1673|1673]]</sup> ). At present, there is limited evidence about the conditions under which EbA measures can self-adapt and when they require human intervention to recover.

<span id="section-2"></span>
<!-- START TABLE -->
'''Table 4.8:''' Costs of ecosystem-based adaptation (EbA). MPA is marine protected area.

<!-- TABLE -->
{| class="wikitable"
|-
| Type of measure
| Capital Costs
| Maintenance Costs
|-
| Wetland Conservation
| No data available
| Thinning, clearing debris after storms, etc.: Mangrove: 5000 USD ha –1 yr –1 in Florida (Lewis, 2001) to 11,000 ha –1 yr –1 (Aerts, 2018).
For mangroves globally, 7–85 USD ha –1 yr –1 (Aerts et al., 2018a);

For marshes in the Wadden Sea, 25 USD m –1 yr –1 (Vuik et al., 2019).

|-
| Wetland Restoration (Marshes/Mangroves, Maritime Forests)
| Wetlands: 85,000 – 230,000 USD ha –1 (Aerts et al., 2018a); Mangroves: USD 9000 ha –1 (median; Bayraktarov et al., 2016);
2000 – 13,000 USD ha –1 in American Samoa (Gilman and Ellison, 2007); Salt Marshes: 67,000 USD ha –1 (Bayraktarov et al., 2016); Brushwood dams for marsh restoration 150 m –1 (Vuik et al., 2019).

| Similar to maintenance costs for Wetland Conservation
|-
| Reef Conservation (Coral/ Oyster)
| For example, start-up costs for Reef MPAs: 96 – 40,000 USD km -2 (McCrea-Strub et al., 2011).
| For MPAs, 12 million USD yr -1 for the Great Barrier Reef (Balmford et al., 2004).
|-
| Reef Restoration (Coral/ Oyster)
| 165,600 USD ha –1 (median; Bayraktarov et al., 2016); Oyster Reefs: 66,800 USD ha –1 (median; Bayraktarov et al., 2016); Artificial Reefs in the UK 30,000–90,000 USD 100 m –1 (Aerts et al., 2018a)
| Similar to maintenance costs for Reef Conservation
|}
<!-- END TABLE -->

<div id="section-4-4-2-3ecosystem-based-adaptation-block-4"></div>

<span id="effectiveness-of-ecosystem-based-adaptation"></span>
===== 4.4.2.3.4 Effectiveness of ecosystem-based adaptation =====

While EbA has been able to reduce the impacts of sea level related hazards, there is still ''little agreement'' on the size of the effect (Gedan et al., 2011 <sup>[[#fn:r1674|1674]]</sup> ; Doswald et al., 2012 <sup>[[#fn:r1675|1675]]</sup> ; Lo, 2016; Renaud et al., 2016 <sup>[[#fn:r1676|1676]]</sup> ). Dozens of independent field, experimental and numerical studies have observed and measured the wave attenuation and flood reduction benefits provided by natural habitats, such as marsh and mangrove wetlands (Barbier and Enchelmeyer, 2014 <sup>[[#fn:r1677|1677]]</sup> ; Möller et al., 2014 <sup>[[#fn:r1678|1678]]</sup> ; Rupprecht et al., 2017 <sup>[[#fn:r1679|1679]]</sup> ), coral reefs (Ferrario et al., 2014 <sup>[[#fn:r1680|1680]]</sup> ; Storlazzi et al., 2017 <sup>[[#fn:r1681|1681]]</sup> ), oyster reefs (Scyphers et al., 2011 <sup>[[#fn:r1682|1682]]</sup> ) and submerged seagrass beds (Infantes et al., 2012 <sup>[[#fn:r1683|1683]]</sup> ). Local and global numerical studies indicate that marshes and mangroves can reduce present-day surge-related flood damages by &gt;15% annually, and the loss of a metre of living coral reef can double annual wave-related flood damages (Narayan et al., 2017 <sup>[[#fn:r1684|1684]]</sup> ; Beck et al., 2018 <sup>[[#fn:r1685|1685]]</sup> ). Artificial reef restoration along tens of metres of coastline using Reef Ball™ and other structures has been shown to reduce wave heights and stabilise beach widths (Reguero et al., 2018a <sup>[[#fn:r1686|1686]]</sup> ; Torres-Freyermuth et al., 2018 <sup>[[#fn:r1687|1687]]</sup> ). The effectiveness of EbA measures, however, varies considerably depending on storm, wetland, reef and landscape parameters (Koch et al., 2009 <sup>[[#fn:r1700|1700]]</sup> ; Loder et al., 2009 <sup>[[#fn:r1701|1701]]</sup> ; Wamsley et al., 2010 <sup>[[#fn:r1702|1702]]</sup> ; Pinsky et al., 2013 <sup>[[#fn:r1703|1703]]</sup> ; Quataert et al., 2015 <sup>[[#fn:r1704|1704]]</sup> ), which makes it difficult to extrapolate the physical and economic benefits across geographies. Depending on these parameters, rates of surge attenuation can vary between 5–70 cm km <sup>-1</sup> (Krauss et al., 2009 <sup>[[#fn:r1705|1705]]</sup> ; Vuik et al., 2015 <sup>[[#fn:r1706|1706]]</sup> ).

Critical gaps remain in our understanding about those parameters that together affect the success of ecosystem-based measures including choice of species and restoration techniques, lead time, natural variability and residual risk, temperature, salinity, wave energy and tidal range (Smith, 2006 <sup>[[#fn:r1707|1707]]</sup> ; Stiles Jr, 2006 <sup>[[#fn:r1708|1708]]</sup> ). Among reasons commonly cited for the failure of mangrove restoration projects are poor choice of mangrove species, planting in the wrong tidal zones and in areas of excessive wave energy (Primavera and Esteban, 2008 <sup>[[#fn:r1709|1709]]</sup> ; Bayraktarov et al., 2016 <sup>[[#fn:r1710|1710]]</sup> ; Kodikara et al., 2017 <sup>[[#fn:r1711|1711]]</sup> ).

The effectiveness of ecosystem-based measures also exhibits high seasonal, annual and longer-term variability. For example, marsh and seagrass wetlands typically have lower densities in winter which reduces their coastal protection capacity (Möller and Spencer, 2002 <sup>[[#fn:r1712|1712]]</sup> ; Paul and Amos, 2011 <sup>[[#fn:r1713|1713]]</sup> ; Schoutens et al., 2019 <sup>[[#fn:r1714|1714]]</sup> ). In the long-term, there is ''limited evidence'' and ''low agreement'' on how changes in sea level, sediment inputs, ocean temperature and ocean acidity will influence the extent, distribution and health of marsh and mangrove wetlands, coral reefs and oyster reefs (Hoegh-Guldberg et al., 2007 <sup>[[#fn:r1715|1715]]</sup> ; Lovelock et al., 2015 <sup>[[#fn:r1716|1716]]</sup> ; Crosby et al., 2016 <sup>[[#fn:r171|171]]</sup> ; Albert et al., 2017 <sup>[[#fn:r1718|1718]]</sup> ). EbA measures may have differential lead times before they are effective. For example, newly planted mangroves provide less wave attenuation until they mature (~3–5 years; Mazda et al., 1997 <sup>[[#fn:r1719|1719]]</sup> ). In contrast, a reef restoration project that uses submerged concrete structures performs as a breakwater as soon as the sub-structure is in place (Reguero et al., 2018a <sup>[[#fn:r1720|1720]]</sup> ).

<div id="section-4-4-2-3ecosystem-based-adaptation-block-5"></div>

<span id="co-benefits-and-drawbacks-of-ecosystem-based-adaptation"></span>
===== 4.4.2.3.5 Co-benefits and drawbacks of ecosystem-based adaptation =====

There is high confidence that ecosystem-based measures provide multiple co-benefits such as sequestering carbon (Siikamäki et al., 2012 <sup>[[#fn:r1721|1721]]</sup> ; Hamilton and Friess, 2018 <sup>[[#fn:r1722|1722]]</sup> ), income from tourism (Carr and Mendelsohn, 2003 <sup>[[#fn:r1723|1723]]</sup> ; Spalding et al., 2017 <sup>[[#fn:r1724|1724]]</sup> ), enhancing coastal fishery productivity (Carrasquilla-Henao and Juanes, 2017 <sup>[[#fn:r1725|1725]]</sup> ; Taylor et al., 2018 <sup>[[#fn:r1726|1726]]</sup> ), improving water quality (Coen et al., 2007 <sup>[[#fn:r1727|1727]]</sup> ; Lamb et al., 2017 <sup>[[#fn:r1728|1728]]</sup> ), providing raw material for food, medicine, fuel and construction (Hussain and Badola, 2010 <sup>[[#fn:r1729|1729]]</sup> ; Uddin et al., 2013 <sup>[[#fn:r1730|1730]]</sup> ), and a range of intangible and cultural benefits (Scyphers et al., 2015 <sup>[[#fn:r1731|1731]]</sup> ) that help improve the resilience of communities vulnerable to sea level hazards (Sutton-Grier et al., 2015 <sup>[[#fn:r1732|1732]]</sup> ).

In comparison to hard structures like seawalls, EbA measures, particularly coastal wetlands, require more land (The Royal Society Science Policy Centre, 2014), and competition for land is often why the ecosystems have declined in the first place (4.4.2.3.1). On developed coasts, this land is often not available. In such cases, hybrid measures that either combine EbA measures with structural measures like mangrove forests in front of dikes (Dasgupta et al., 2019 <sup>[[#fn:r1733|1733]]</sup> ), or build ecological enhancements into engineered structures can provide an effective solution. Like any other feature that interacts with coastal processes, natural wetlands and reefs can increase flooding in some instances, for example, due to the redistribution or acceleration of flows in channels within a wetland system (Marsooli et al., 2016 <sup>[[#fn:r1734|1734]]</sup> ), or an increase in infragravity wave (i.e., surface gravity waves with frequencies lower than wind waves) energy behind a reef (Roeber and Bricker, 2015 <sup>[[#fn:r1735|1735]]</sup> ).

<div id="section-4-4-2-3ecosystem-based-adaptation-block-6"></div>

<span id="governance-of-ecosystem-based-adaptation"></span>
===== 4.4.2.3.6 Governance of ecosystem-based adaptation =====

The coastal protection benefits of natural ecosystems are increasingly being recognised within international discourse and national coastal adaptation, resilience and sustainable development plans and strategies (Section 4.4.2.3.1). In general, obtaining permits for EbA remains more difficult compared to established hard protection measures, in places like the USA (Bilkovic, 2017 <sup>[[#fn:r1736|1736]]</sup> ). However, there are examples of instruments specifically tailored to retain the protective function of EbA (Borges et al., 2009 <sup>[[#fn:r1737|1737]]</sup> ; Government of India, 2018 <sup>[[#fn:r1738|1738]]</sup> ). The Living Shorelines Regulations of the state government of Maryland in the USA (Maryland DEP, 2013 <sup>[[#fn:r1739|1739]]</sup> ), for instance, requires that private properties must include marsh creation or other non-structural measures when stabilising their shorelines, unless a waiver is obtained.

There are an increasing number of public and private financial mechanisms and policy instruments to encourage the use and implementation of EbA measures (Colgan et al., 2017 <sup>[[#fn:r1740|1740]]</sup> ; Sutton-Grier et al., 2018 <sup>[[#fn:r1741|1741]]</sup> ). For example, a regulation by the Federal Emergency Management Agency (FEMA) of the USA, allows proponents of hazard mitigation projects, such as state, territorial and local governments, to take into account the co-benefits of EbA when assessing benefit-cost ratios of FEMA-funded recovery projects (FEMA, 2015 <sup>[[#fn:r1742|1742]]</sup> ). International guidelines are being developed for designing and implementing EbA measures, with the intention to support wider implementation of these responses (Hardaway Jr and Duhring, 2010 <sup>[[#fn:r1743|1743]]</sup> ; Van Slobbe et al., 2013; Van Wesenbeeck et al., 2017; Bridges et al., 2018 <sup>[[#fn:r1744|1744]]</sup> ).

<div id="section-4-4-2-3ecosystem-based-adaptation-block-7"></div>

<span id="economic-efficiency-of-ecosystem-based-adaptation"></span>
===== 4.4.2.3.7 Economic efficiency of ecosystem-based adaptation =====

There is ''limited evidence'' regarding the economic efficiency of EbA, mainly due to the ''low agreement'' about EbA effectiveness (Section 4.4.2.3.2) and costs (Section 4.4.2.3.2). A study of coastal protection measures on the Gulf of Mexico coastline, USA, estimated that EbA measures have average benefit-cost ratios above 3.5 for 2030 flood risk conditions, assuming a discount rate of 2% (Reguero et al., 2018b <sup>[[#fn:r1745|1745]]</sup> ; see Section 4.4.2.3.2). This study also finds that EbA are nearly four times more cost-efficient along developed coastlines as compared to conservation-priority areas because protection benefits are higher in the former case due to the level of asset exposure.

<div id="section-4-4-2-4advance"></div>

<span id="advance"></span>
==== 4.4.2.4 Advance ====

<div id="section-4-4-2-4advance-block-1"></div>

<span id="observed-advance-across-geographies"></span>
===== 4.4.2.4.1 Observed advance across geographies =====

<div id="section-4-4-2-4advance-block-2"></div>

Advance has a long history in most areas where there are dense coastal populations and a shortage of land ( ''very high confidence'' ). This includes land reclamation through polders around the southern North Sea (Germany, the Netherlands, Belgium and England) and China (Wang et al., 2014), which coincides with regions where there is extensive hard protection in place (Section 4.4.2.4). Land reclamation has also taken place in all major coastal cities to some degree, even if only for the creation of port and harbour areas by raising coastal flats above normal tidal levels through sediment infill. On some steep coasts, where there is little flat land, such as the Hong Kong Special Administrative Region of China, material from elevated areas has been excavated to create fill material to build land out into the sea.

Globally, it is estimated that about 33,700 km <sup>2</sup> of land has been gained from the sea during the last 30 years (about 50% more than has been lost), with the biggest gains being due to land reclamation in places like Dubai, Singapore and China (Wang et al., 2014; Donchyts et al., 2016). In Shanghai alone, 590 km <sup>2</sup> land has been reclaimed during the same period (Sengupta et al., 2018). In Lagos, 25 km² of new land is currently being reclaimed (www.ekoatlantic.com). Land reclamation is also popular in some small island settings. The Maldives has recently increased the land area of their capital region by constructing a new island called Hulhumalé, which has been built 60 cm higher than the normal island elevation of 1.5 m, in order to take into account future SLR (Hinkel et al., 2018).

<div id="section-4-4-2-4advance-block-3"></div>

<span id="projected-advance"></span>
===== 4.4.2.4.2 Projected advance =====

<div id="section-4-4-2-4advance-block-4"></div>

Advance was not primarily a response to SLR in the past, but due to a range of drivers, including land scarcity, population pressure and extreme events, future advance measures are expected to become more integrated with coastal adaptation and might even be seen as an opportunity to support and fund adaptation in some cases (Linham and Nicholls, 2010; RIBA and ICE, 2010; Nicholls, 2018). While there is no literature on this topic, significant further advance measures can be expected in land scarce situations, such as found in China, Japan and Singapore, in coming decades.

<div id="section-4-4-2-4advance-block-5"></div>

<span id="costs-of-advance"></span>
===== 4.4.2.4.3 Costs of advance =====

<div id="section-4-4-2-4advance-block-6"></div>

Contrary to protection measures, little systematic monetary information is available about costs of advance measures, specifically not in the peer reviewed literature. The costs of land reclamation are extremely variable and depend on the unit cost of fill versus the volumetric requirement to raise the land. Hence, filling shallow areas is preferred on a cost basis.

<div id="section-4-4-2-4advance-block-7"></div>

<span id="effectiveness-of-advance"></span>
===== 4.4.2.4.4 Effectiveness of advance =====

<div id="section-4-4-2-4advance-block-8"></div>

Similar to hard protection, land reclamation is mature and effective technology and can provide predictable levels of safety. If the entire land area is raised above the height of ESLs, residual risks are lower as compared to hard protection as there is no risk of catastrophic defence failure.

<div id="section-4-4-2-4advance-block-9"></div>

<span id="co-benefits-and-drawbacks-of-advance"></span>
===== 4.4.2.4.5 Co-benefits and drawbacks of advance =====

<div id="section-4-4-2-4advance-block-10"></div>

The major co-benefit of advance is the creation of new land. The major drawbacks include groundwater salinisation, enhanced erosion and loss of coastal ecosystems and habitat, and the growth of the coastal floodplain (Li et al., 2014; Nadzir et al., 2014; Wang et al., 2014; Chee et al., 2017). In China, for example, about 50% of coastal ecosystems have been lost due to land reclamation, leading to a range of impacts such as loss of biodiversity, decline of bird species and fisheries resources, reduced water purification, and more frequent harmful algal blooms (Wang et al., 2014). For example, the reclamation of about 29,000 ha of land in Saemangeum, Republic of Korea, in 2006, has led to a decrease in shorebird numbers by over 30% in two years, probably caused by mortality (Moores et al., 2016). Inadvertently, historic land reclamation through polderisation may have enhanced exposure and risk to coastal flooding by creating new populated floodplains, but this has not been evaluated.

<div id="section-4-4-2-4advance-block-11"></div>

<span id="governance-of-advance"></span>
===== 4.4.2.4.6 Governance of advance =====

<div id="section-4-4-2-4advance-block-12"></div>

Land reclamation raises equity issues with regards to access and distribution of the new land created, specifically due to the political economy associated with high coastal land values, and the involvement of private capital and interests (Bisaro and Hinkel, 2018), but this has hardly been explored in the literature.

<div id="section-4-4-2-4advance-block-13"></div>

<span id="economic-efficiency-of-advance"></span>
===== 4.4.2.4.7 Economic efficiency of advance =====

<div id="section-4-4-2-4advance-block-14"></div>

There is ''limited evidence'' on the efficiency of advance responses in the scientific literature. Benefit-cost ratios of land reclamation can be very high in urban areas due to high land and real estate prices (Bisaro and Hinkel, 2018).

<div id="section-4-4-2-5accommodation"></div>

<span id="accommodation"></span>
==== 4.4.2.5 Accommodation ====

<div id="section-4-4-2-5accommodation-block-1"></div>

<span id="observed-accommodation-across-geographies"></span>
===== 4.4.2.5.1 Observed accommodation across geographies =====

<div id="section-4-4-2-5accommodation-block-2"></div>

There is a ''high agreement'' that accommodation is a core element of adaptation, and it is taking place on various scales based on measures such as flood proofing and raising buildings, implementing drainage systems, land use changes as well as EWS, emergency planning, setback zones and insurance schemes. However, no literature is available that summarises observed accommodation worldwide. There is ''low evidence'' of accommodation occurring directly as a consequence of SLR but ''high evidence'' of accommodation measures being implemented in response to coastal hazards such as coastal flooding, salinisation and other sea-borne hazards such as cyclones.

Flood proofing may include the use of building designs and materials which make structures less vulnerable to flood damages and/or prevent floodwaters from entering structures. Examples include floating houses in Asia, such as in Vietnam (Trang, 2016), raising the floor of houses in the lower Niger delta (Musa et al., 2016), construction of verandas with sandbags and shelves in houses to elevate goods during floods in coastal communities in Cameroon (Munji et al., 2013). In Semarang City, Indonesia, residents adapted to coastal flooding by elevation of their houses by 50–400 cm or by moving their goods to safer places, without making structural changes (Buchori et al., 2018). Residents of Can Tho City of the Mekong Delta, Vietnam elevated houses in response to tidal flooding (Garschagen, 2015). In urban areas extensive drainage systems contribute to accommodation such as in Hong Kong and Singapore, which rely on urban drainage systems to handle large volumes of surface runoff generated during storm events (Chan et al., 2018). Farming practices have been adapted to frequent flooding in the lower Niger delta: farmers raise crops above floodwaters by planting on mounds of soil and apply ridging and terracing on farmlands to form barriers (Musa et al., 2016). In the floodplains of Bangladesh, floating gardens help to maintain food production even if the area is submerged (Irfanullah et al., 2011). Here, the traditional way to build homesteads is on a raised mound, built with earth from the excavation of canals and ponds (ADPC, 2005). Coastal infrastructure, such as ports, having a functional need to be at the coast, accommodate SLR with elevated piers and critical infrastructure. One example is Los Angeles, where PierS was raised to an elevation of 6 m (Aerts, 2018).

Communities in the Netherlands are experimenting with floating/amphibious houses capable of adapting to different water levels, and similar considerations are also discussed in other geographies, such as in Bangkok (Nilubon et al., 2016). Flood proofing is widely applied in the USA, where wet and dry flood proofing measures are recognised: wet flood proofing reduces damage from flooding while dry flood proofing makes a building watertight or substantially impermeable to floodwaters up to the expected flood height (FEMA, 2014). In that sense, dry flood proofing could also be interpreted as a protection measure on the level of individual structures.

Physical accommodation to salinisation and saline water intrusion is more poorly documented. It mainly entails agricultural adaptation to soil salinity, and saline surface and ground water, as described for the land use changes aimed at alternating rice-shrimp systems and shrimp aquaculture in the Mekong Delta (Renaud et al., 2015) or using methods which decrease soil salinity, such as flushing rice fields with fresh water to wash out salinity (Renaud et al., 2015), or applying maize straw in wheat fields (Xie et al., 2017). Coastal communities are also experimenting with the use of salt tolerant varieties as a result of breeding programmes, for example, in Indonesia (Rumanti et al., 2018), or saline irrigation water in conjunction with fresh water, such as for maize in coastal Bangladesh (Murad et al., 2018).

Adaptation planning for SLR has been incorporated into land use planning in several states in the USA (Butler et al., 2016b). In the Yangtze River Delta, landscape planning designs floodplain zones to accept floodwaters (Seavitt, 2013). In the Mekong Delta, different land use options, including shifting from freshwater agriculture to brackish and saline agriculture, were proposed as seawater intrudes farther inland (Smajgl et al., 2015).

EWS are frequently incorporated into overall risk reduction strategies and are applied for various coastal hazards such as tsunamis in coastal areas of Indonesia (Lauterjung et al., 2017) and hydro-meteorological coastal hazards in Bangladesh and Uruguay (Leal Filho et al., 2018). They fall under ‘accommodation’ as they allow people to remain in the hazard-prone area but provide advance warning or evacuation in the face of imminent danger. In contrast to hard protection measures, EWS have shorter installation time and lower impact on the environment (Sättele et al., 2015). They can work effectively to reduce risk arising from predictable hazardous events but are less well-suited to accommodate slow onset change (i.e., events or processes that happen with high certainty under different climate change scenarios)

Climate risk insurance schemes have been recently developed to address sudden, and in rare cases, slow onset hazards at the coast, and to increase overall resilience. For coastal risks, insurance is mainly applicable for sudden onset hazards, including storm surges and coastal flooding, to buffer against the financial impacts of loss events. For slow onset hazards, insurance schemes are not the first-best tool, whereas resilience building and prevention of loss and damage in such instances may be more cost-effective ways to address these risks (Warner et al., 2013). In this context, index based insurance products are increasingly offered, particularly in low-income countries and have also been included in a number of countries in their NDCs and in some cases in their National Adaptation Plans (NAPs; Kreft et al., 2017). Countries with existing climate risk insurance schemes include, for example, Haiti, Maldives, Seychelles and Vietnam. The InsuResilience Global Partnership for Climate and Disaster Risk Finance and Insurance Solutions was launched at the 2017 UN Climate Conference (COP 23) in Bonn. InsuResilience aims to enable more timely response after a disaster and helps to better prepare for climate and disaster risk through the use of climate and disaster risk finance and insurance solutions. So far, climate risk insurance was used mainly in the context of agriculture, where it has showed great efficacy in boosting investments for increasing productivity (Fernandez and Schäfer, 2018). However, on the global scale, the uptake of index insurance is still low (Yuzva et al., 2018).

<div id="section-4-4-2-5accommodation-block-3"></div>

<span id="projected-accommodation"></span>
===== 4.4.2.5.2 Projected accommodation =====

<div id="section-4-4-2-5accommodation-block-4"></div>

While there is no literature on projected accommodation, current trends suggest further uptake of accommodation approaches in coming decades, especially where protection approaches are not economically viable. Flood proofing of houses and establishment of new building codes to accommodate coastal hazards is also expected to become more common in coming decades. Similarly, accommodation measures for salinity are under further development, such as rice breeding programs to improve salt tolerance (Linh et al., 2012; Quan et al., 2018b). However, the achievements to improve salinity tolerance in rice are rather modest so far (Hoang et al., 2016) although efforts are expected to continue or even intensify. Given that index based insurance products have been included in NDCs and NAPs in a number of countries (Kreft et al., 2017), uptake is expected to grow. Ports can continue elevating hazard-prone facilities and the critical parts of port infrastructure can be protected by flood walls. Alternatively, ports can use advance measures to develop port facilities seaward (Aerts, 2018).

In summary, due to the large variety of different measures implemented in ad hoc ways worldwide, there is ''low confidence'' in quantitative projections of accommodation measures in response to SLR. However, there is ''high confidence'' that accommodation measures will continue to be a widespread adaptation option especially in combination with protection and retreat measures.

<div id="section-4-4-2-5accommodation-block-5"></div>

<span id="cost-of-accommodation"></span>
===== 4.4.2.5.3 Cost of accommodation =====

<div id="section-4-4-2-5accommodation-block-6"></div>

The cost of accommodation varies widely with the measures taken as well as the expected flood height. For flood proofing of buildings in New York City for instance, Aerts et al. (2014) provided an economic rationale for the implementation of improved building codes, such as elevating new buildings and protecting critical infrastructure (see also Box 4.1). Flood proofing can also be undertaken by individuals and even small, inexpensive flood proofing efforts can result in reductions in flood damage (Zhu et al., 2010). In general, costs for flood proofing increase as the flood protection elevation increases. Other costs include those for maintenance and, if applicable, insurance premiums. For example, deciding to elevate a building in the USA will increase the project’s cost; however, the additional elevation may lead to significant savings on flood insurance premiums (FEMA, 2014).

<div id="section-4-4-2-5accommodation-block-7"></div>

<span id="effectiveness-of-accommodation"></span>
===== 4.4.2.5.4 Effectiveness of accommodation =====

<div id="section-4-4-2-5accommodation-block-8"></div>

Accommodation measures can be very effective for current conditions and small amounts of SLR, also buying time to prepare for future SLR. Success stories include the case of Bangladesh where improved early warnings, the construction of shelters, and development of evacuation plans, helped to reduce fatalities as a result of flooding and cyclones (Haque et al., 2012). Illiteracy, lack of awareness and poor communication are, however, still hampering the effectiveness of early warnings (Haque et al., 2012). If well designed, and if the premiums reflect individual risks, insurance can effectively discourage further investments in risky areas as insurance cost provides information on the nature of locality-specific risks and can incentivise investment in risk reduction by requiring that certain minimum standards are met before granting insurance coverage (Kunreuther, 2015). Limits to such accommodation occur much earlier compared to protect, advance and retreat measures. While dikes can be raised to 10 m, and retreat can be implemented to the 10 m contour or higher, accommodating SLR has practical and economic limits, and ultimately retreat or protection will be required.

<div id="section-4-4-2-5accommodation-block-9"></div>

<span id="co-benefits-and-drawbacks-of-accommodation"></span>
===== 4.4.2.5.5 Co-benefits and drawbacks of accommodation =====

<div id="section-4-4-2-5accommodation-block-10"></div>

The major co-benefit of accommodation is improved resilience of ''in situ'' communities without retreat or the use of land and resources for the construction of protection measures. Flood proofing, for example, helps prevent demolition or relocation of structures and it is often an affordable and cost effective approach to reducing flood risk (Zhu et al., 2010). Specific accommodation measures have different co-benefits such as that stilt houses not only protect from flooding but also from wild animals (Biswas et al., 2015). Accommodation—depending on the measure implemented—has the potential to maintain landscape connectivity allowing access to the ocean as well as landward migration of ecosystems, at least to some degree. It also retains flood dynamics and with that the benefits of flooding such as sediment re-distribution. Stilt houses leave space for the floodwater while wet-flood proofing maintains a low hydrostatic pressure on the buildings so that structures are less prone to failure during flooding (FEMA, 2014)

The major drawback of accommodation is that it actually does not prevent flooding or salinisation, which might have consequences not addressed by the accommodation measure itself. Examples include inundation of an area where houses are flood proofed but schooling of children and business operations are nevertheless disrupted. Significant clean up may also be needed after flood water enters buildings, including the removal of sediment, debris or chemical residues (FEMA, 2014). Also, flood proofing measures require the current risk of flooding to be known and communicated to and understood by the public through flood hazard mapping studies and flood warning information (Zhu et al., 2010). Small businesses in particular may face difficulties to recover from flooding due to lack of forward planning (Hoggart et al., 2014).

Co-benefits of insurance include the possibility that sovereign level insurance may improve the credit ratings of vulnerable countries, reducing the cost of capital and allowing them to borrow to invest in resilient infrastructure (Buhr et al., 2018). Major natural disasters can weaken sovereign ratings, especially if there is no insurance in place (Moritz Kramer, 2015). One much discussed drawback of insurance is the moral hazard that may result: since someone else bears the costs of a loss, those insured may be less inclined to take precautionary measures or may act recklessly (Duus-Otterström and Jagers, 2011).

<div id="section-4-4-2-5accommodation-block-11"></div>

<span id="governance-of-accommodation"></span>
===== 4.4.2.5.6 Governance of accommodation =====

<div id="section-4-4-2-5accommodation-block-12"></div>

While accommodation measures to coastal hazards are often taking place at the local level, and are decided by individual homeowners, farmers or communities, from a governance perspective it is important to provide guidance on how and to what extent owners can retrofit their homes to reduce the risk to coastal flooding. In New York City, for instance, changes to building codes, require elevating, or flood proofing of existing and new buildings in the 100-year floodplain, and prevent construction of critical infrastructure like hospitals in the flood zone (NYC, 2014; see also Box 4.1).

<div id="section-4-4-2-5accommodation-block-13"></div>

Effective coastal risk management efforts rely on good governance that includes understanding the probability and consequences of hazard impacts like flooding and salinisation, and implementing mechanisms to prevent or manage all possible events (EEA, 2013). The effectiveness of accommodation measures based on institutional measures, such as EWS and evacuation plans, largely depends on the governance capabilities they are embedded in.

<div id="section-4-4-2-5accommodation-block-14"></div>

<span id="economic-efficiency-of-accommodation"></span>
===== 4.4.2.5.7 Economic-efficiency of accommodation =====

<div id="section-4-4-2-5accommodation-block-15"></div>

There is ''high confidence'' that many accommodation measures are very cost-efficient. Flood EWS coupled with precautionary measures have been shown to produce significant economic benefits (Parker, 2017). Elevating areas at high risk and retrofitting buildings in Ho Chi Minh City, for example, have benefit-cost ratios of 15 under SLR of 180 cm and a discount rate of 5% during the 21st century (Scussolini et al., 2017). In the context of the National Flood Insurance Program in the USA, it has been estimated that elevating new houses by 60 cm might raise mortgage payments by 240 USD yr <sup>-1</sup> , but reduce flood insurance by 1000–2000 USD yr <sup>-1</sup> depending on the flood zone (FEMA, 2018), although this only addresses present extremes and ignores future SLR (Zhu et al., 2010). In Europe, the benefits of installing a cross-border continental-scale flood EWS are estimated at 400 EUR per EUR invested (Pappenberger et al., 2015).

<div id="section-4-4-2-6retreat"></div>

<span id="retreat"></span>
==== 4.4.2.6 Retreat ====

<div id="section-4-4-2-6retreat-block-1"></div>

<span id="observed-retreat-across-geographies"></span>
===== 4.4.2.6.1 Observed retreat across geographies =====

There is ''limited evidence'' of migration occurring directly as a consequence of impacts associated with environmental change generally and SLR specifically. Research examining the linkages between migration and environmental change has been conducted in the Pacific (Connell, 2012; Janif et al., 2016; Perumal, 2018), South Asia (Szabo et al., 2016; Call et al., 2017; Stojanov et al., 2017), Latin America (Nawrotzki and DeWaard, 2016; Nawrotzki et al., 2017), Alaska, in North America (Marino and Lazrus, 2015; Hamilton et al., 2016) and Africa (Gray and Wise, 2016). While some limited evidence was found on population movement inland associated with shoreline encroachment in Louisiana, USA (Hauer et al., 2018), this research emphasises that the relationship between climate change impacts including SLR and migration is more nuanced than suggested by simplified cause-and-effect models (Adger et al., 2015). Migration is driven by a large number of individual, social, economic, political, demographic and environmental push and pull factors (Black et al., 2011; Koubi et al., 2016), interwoven with mega-trends such as urbanisation, land use change and globalisation, and is influenced by development and political practices and discourses (Bettini and Gioli, 2016; Cross-Chapter Box 7). For example, asset endowed individuals and households are more able to migrate out from flood-prone areas (Milan and Ruano, 2014; Logan et al., 2016), while the poorest households are significantly susceptible to material and human losses following an extreme event or disruptive environmental change (Call et al., 2017). Individual and social drivers include perceptions of environmental change (Koubi et al., 2016), formed by both direct experience of change and indirect information from social networks, mass media and governmental agencies. Environmental factors include the longer term impacts of climate variability and change, which can erode the capacity of ecosystems to provide essential services such as availability of freshwater, soil fertility and energy production acting as a threat multiplier for other drivers of migration(Hunter et al., 2015; McLeman, 2018).

There is ''robust evidence'' of disasters displacing people worldwide, but ''limited evidence'' that climate change or SLR is the direct cause. In 2017, 18.8 million people were displaced by disasters, of which 18 million were displaced by weather-related events including 8.6 million people displaced by floods and 7.5 million by storms, with hundreds of millions more at risk (IDMC, 2017; Islam and Khan, 2018). The majority of resultant population movements tend to occur within the borders of affected countries (Warner and Afifi, 2014; Hunter et al., 2015; Nawrotzki et al., 2017).

We find ''robust evidence'' of planned relocation taking place worldwide in low-lying zones exposed to the impacts of coastal hazards (Hino et al., 2017; Mortreux et al., 2018). While relocation plans are usually discussed after an extreme event occurs, they generally target the reduction of long-term environmental risks, including those of SLR (McAdam and Ferris, 2015; Hino et al., 2017; Morrison, 2017). For example, in the aftermath of Hurricane Katrina, the Louisiana Comprehensive Master Plan for a Sustainable Coast recommended the relocation of several communities in the next 50 years due to expected RSL rise, and relocation of inhabitants from Isle de Jean Charles is already taking place (Barbier, 2015; Coastal Protection and Restoration Authority of Louisiana, 2017). In Shismaref, an Iñupiat community in Alaska, increased shoreline erosion triggered government-led relocation (Bronen and Chapin, 2013; Maldonado et al., 2013). In the Pacific, current coastal risks aggravated by rising sea level are driving the government led relocation of the inhabitants of Taro, the provincial capital of Choiseul Province in the Solomon Islands (Albert et al., 2018). In 2014, the government of Kiribati purchased land on Vanua Levu, the second largest island of Fiji, with the purpose of economic development and food security, but many i-Kiribati associated the acquisition with future relocation to Fiji (Hermann and Kempf, 2017). In southeast Asia, the government of Vietnam assists and manages rural populations’ relocation from disaster prone areas exposed to coastal risks in the Mekong Delta to large industrial areas with high labour demand, such as Ho Chi Minh City and Can Tho City (Collins et al., 2017). Managed realignment carried out for the purposes of habitat creation, improved flood risk management and more affordable coastal protection, is increasingly popular in Europe, but usually involves small-scale projects and few people if any (Esteves, 2013). Most of the managed realignment projects in the UK and Germany have been carried out for habitat creation and to reduce spending on coastal defences (Hino et al., 2017)

<div id="section-4-4-2-6retreat-block-2"></div>

<span id="projected-retreat"></span>
===== 4.4.2.6.2 Projected retreat =====

There is ''high agreement'' that climate change has the potential to drastically alter the size and direction of migration flows (Connell, 2012; Gray and Wise, 2016; Janif et al., 2016; Nawrotzki and DeWaard, 2016; Szabo et al., 2016; Call et al., 2017; Nawrotzki et al., 2017), but there is ''low confidence'' in quantitative projections of migration in response to SLR and extremes of sea level. The number of modelling studies of migration in response to environmental drivers has increased rapidly over the past decade (Kumari et al. 2018), but only a small portion of these model studies address migration in response to SLR and sea level extremes. Amongst these, a variety of different modelling approaches have been applied, but no model currently accounts for all push and pull factors influencing migration decisions (see Section 4.4.2.6.1).

A model projecting future US county-level populations exposed to permanent inundation was combined with an empirical model of potential migration destinations to produce the first sea level/migration analysis of migrant destinations (Hauer, 2017). Assuming that households with incomes above 100,000 USD yr <sup>-1</sup> would have resources to stay and adapt, it was found that 1.8 m SLR by 2100 would displace over two million people in south Florida. Projected population gains due to SLR reach several hundred thousand for some inland urban areas. A gravity model modified to account for both distance to destinations and their attractiveness (deriving from such factors as economic opportunity and environmental amenities) projects a net migration into and out of the East African coastal zone, ranging from out-migration of 750,000 people between 2020 and 2050 to a small in-migration (Kumari et al., 2018). However, this range includes migration stimulated by freshwater availability as well as SLR and episodic flooding. A generalised radiation or diffusion model predicts 0.9 million people will migrate due to SLR in Bangladesh by 2050 and 2.1 million by 2100, largely internally, with substantial implications for nutrition, shelter and employment in destination areas (Davis et al., 2018).

A global dynamic general equilibrium framework (Desmet et al., 2018) provides a more comprehensive approach to accounting for economic factors including changes to trade, innovation, and agglomeration, and political factors, such as policy barriers to mobility, all of which influence the migration response to environmental change. Agent-based models attempt to simulate decisions by individuals who face a variety of socioeconomic and environmental changes (Kniveton et al., 2012). However, neither general equilibrium nor agent-based frameworks have been applied yet to migration responses to SLR. Econometric models, common in climate/migration studies (Millock, 2015), likewise have yet to be applied to the SLR context, except for a single case study where an econometric model was used to interpret the outcome of a discrete choice experiment (Buchanan et al., 2019). For example, an interesting distinction between migration responses to long term temperature and precipitation trends in contrast to extreme events like flooding has been noted (Bohra-Mishra et al., 2014; Mueller et al., 2014), but similar econometric studies have yet to be done comparing responses to gradual land loss versus flooding during ESL events.

<div id="section-4-4-2-6retreat-block-3"></div>

<span id="cost-of-retreat"></span>
===== 4.4.2.6.3 Cost of retreat =====

We have ''limited evidence'' of estimates on the cost of retreat. There are few cost estimates in the literature and these are based on stylised assumptions as little empirical data is available.

The cost of managed relocation, including land acquisition, building of roads and infrastructure and other subsidies, was found to vary from 10,000–270,000 GBP per home in United Kingdom Coastal Change Pathfinder projects (Regeneris Consulting, 2011), and between 10,000 USD in Fiji and 100,000 USD per person in Alaska and in the Isle of Jean Charles in the USA (Hino et al., 2017). For people involved in planned relocation in Shaanxi Province, Northwest China, households receive subsidies ranging from 1200–5100 USD (Lei et al., 2017). The Louisiana’s National Disaster Resilience Competition, Phase II Application states that the proposed relocation of 40 households in the Isle de Jean Charles in Louisiana is estimated to cost 48,379,249 USD, including the cost for land acquisition, infrastructure and construction of new dwellings (State of Louisiana, 2015). Generally, maintenance costs do not arise if people are moved completely out of the hazard zone (Suppasri et al., 2015; Hino et al., 2017). In cases in which people are only moved so that short-term but not long-term risk is reduced, follow up costs for further responses will occur.

The individual costs associated with displacement after an environmental disaster are difficult to obtain. In the literature, there are limited estimates of the social costs to residents of Guadeloupe, Saint Croix, St. Thomas, Puerto Rico, and the southeast USAdisplaced after Hurricanes Hugo (1989) and Katrina (2005). A survey conducted across 18 parishes (i.e. counties) in Louisiana in 2006 revealed that non-displaced households had an average income of 36,000 USD compared to an average income of 30,000 USD recorded for displaced households (Hori and Schafer, 2010).

<div id="section-4-4-2-6retreat-block-4"></div>

<span id="effectiveness-of-retreat"></span>
===== 4.4.2.6.4 Effectiveness of retreat =====

There is ''very high confidence'' that retreat is effective in reducing the risks and impacts of SLR as retreat directly reduces exposure of human settlements and activities (Gioli et al., 2016; Shayegh et al., 2016; Hauer, 2017; Morrison, 2017).

<div id="section-4-4-2-6retreat-block-5"></div>

<span id="co-benefits-and-drawbacks-of-retreat"></span>
===== 4.4.2.6.5 Co-benefits and drawbacks of retreat =====

The other outcomes of retreat responses, beyond the one of effectively reducing SLR risks and impacts, are complex and affect both origin and destination. Generally, retreat impacts social networks, access to services and economic and social opportunities, and several well-being indicators (Jones and Clark, 2014; Adams, 2016; Herath et al., 2017; Kura et al., 2017; McNamara et al., 2018). The socioeconomic benefits of migration to individuals and households may include improved access to health and education services, as well as labour markets (Wrathall and Suckall, 2016). Destination areas may gain economically as populations and capital relocate and provide a new source of labour, capital and innovation to inland areas (see Section 4.4.2.6.2; de Haas, 2010). Income inequality may be reduced, but only through migration to areas with growing economies. Remittances can provide flexibility in livelihood options, supply capital for investment and spread risk (Scheffran et al., 2012).

Drawbacks of migration and displacement at the destination can be increased competition for resources and within labour markets, pressure on frontline services and on social cohesion as a result of heightened cultural or ethnic tension (Werz and Hoffman, 2015), as well as cultural, social and psychological losses related to disruptions to sense of place and identity, self-efficacy, and rights to ancestral land and culture (McNamara et al., 2018). The unplanned and unassisted voluntary relocation of the inhabitants of Nuatambu and Nusa Hope in the Solomon Islands to areas further from the coast poses a series of practical challenges with sanitation, access to drinking water and transport (Albert et al., 2018).

The success of planned relocation in terms of the balance of co-benefits and drawbacks varies across relocation schemes (Hino et al., 2017) and outcomes are highly uneven (Genovese and Przyluski, 2013; Ford et al., 2015; Nordstrom et al., 2015; Bukvic and Owen, 2017; Hino et al., 2017; Jamero et al., 2017). On the one hand, well designed and carefully implemented programmes, such as the ongoing resettlement of indigenous communities in Alaska, can improve housing standards and reduce vulnerability (Suppasri et al., 2015; Albert et al., 2018). On the other hand, relocated communities have often become further impoverished (Wilmsen and Webber, 2015), because they are removed from cultural and material resources on which they rely, compounded by poor implementation processes that may fail to ensure fairness, social and environmental justice and well-being (Herath et al., 2017; Mortreux et al., 2018; Nygren and Wayessa, 2018).

<div id="section-4-4-2-6retreat-block-6"></div>

<span id="governance-of-retreat"></span>
===== 4.4.2.6.6 Governance of retreat =====

Environmentally driven migration and displacement gained major attention over the last decade in the international policy community (Goodwin-Gill and McAdam, 2017). Worldwide programmes, such as the Nansen Initiative, signed by 110 countries to address the serious legal gap around the protection of cross-border migrants impacted by natural disasters, have been implemented (Gemenne and Brücker, 2015). In 2016, the Platform on Disaster Displacement was established to follow up on the work conducted by the Nansen Initiative with the objective of implementing the recommendations of the Protection Agenda (McAdam and Ferris, 2015). Governments are further encouraged by civil society to relocate people at risk and displaced populations out of disaster-prone areas to avoid potential casualties (Lei et al., 2017; Mortreux et al., 2018). There have been discussions among Pacific Island countries and territories and other nations in the Pacific Rim around new policy mechanisms that would facilitate adaptive migration in the region in response to natural hazards including SLR (Burson and Bedford, 2015). There have been cases presented at the Immigration and Protection Tribunal of New Zealand testing refugee claims associated with climate change from Tuvaluan and i-Kiribati applicants, both citing environmental change on their home islands as grounds for remaining in New Zealand. One applicant was successful in the quest to remain in New Zealand on humanitarian grounds, but not on the grounds of refugee status (Farbotko et al., 2016).

The is ''high agreement'' that outcomes can be improved by upholding the principle of procedural justice and respecting the autonomy of individuals and their decisions about where and how they live (Warner et al., 2013; Schade et al., 2015; McNamara et al., 2018). However, there are cases where logistical and political stances constrain the application of such approach, such as when the government of Sri Lanka prohibited rebuilding along the coastline of the country after the 2004 tsunami (Hino et al., 2017). Proactive planning, including participation and consultation with those in peril, has the potential to improve outcomes ( ''medium confidence'' ; de Sherbinin et al., 2011; Gemenne and Blocher, 2017). Governments can assist migrants through policy reforms to enable relocation to fast growing economic regions in the country. An example of this approach was adopted in Vietnam by both the National Target Program to Respond to Climate Change and the National Strategy for Natural Disaster Prevention, Response and Mitigation targeted at locations within the Mekong Delta exposed to the impacts of SLR (Nguyen et al., 2015; Collins et al., 2017). Outcomes of retreat for both community of origin and destination can also be improved by building the human capital of migrants (skills, health and education), reducing costs of migration and remittance transfer, and provision of improved safety nets for migrants at their destinations ( ''high agreement'' ) (Gemenne and Blocher, 2017).

<div id="section-4-4-2-6retreat-block-7"></div>

<span id="economic-efficiency-of-retreat"></span>
===== 4.4.2.6.7 Economic efficiency of retreat =====

There is ''limited evidence'' on the efficiency of retreat responses in the scientific literature.

<span id="governance-challenges-in-responding-to-sea-level-rise"></span>
=== 4.4.3 Governance challenges in responding to sea level rise ===

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

<span id="introduction-2"></span>
==== 4.4.3.1 Introduction ====

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

Governance is pivotal to shaping SLR responses. The assessment of SLR responses above has shown that each type of response raises specific governance challenges associated with the distribution of costs, benefits and negative consequences of responses across societal actors. Hence, SLR responses require governance efforts if social conflicts are to be resolved and mutual opportunities amongst all actors realised. Generally, responses involve the interaction of diverse public and private actors at different levels of decision making with divergent values, interests and goals on coastal activities, lifestyles, livelihoods, risks, resilience and sustainability ( ''high confidence'' ; Dovers and Hezri, 2010; Foerster et al., 2015; Giddens, 2015; Mills et al., 2016; Dolšak and Prakash, 2018; Hinkel et al., 2018; Hoegh-Guldberg et al., 2018; AR5). This leads to a number of overarching governance challenges that arise from the nature of SLR, which will be assessed in this section.

While there is a substantial literature on coastal governance, little attention has been focused explicitly on SLR governance, as was also the case in AR5 (Wong et al., 2014). Furthermore, much of the adaptation governance literature has focused on putting forward normative prescriptions on how governance arrangements ought to be (e.g., transformative governance; Chaffin et al., 2016), but with limited empirical evidence on the actual effectiveness of these prescriptions (Klostermann et al., 2018; Runhaar et al., 2018). Hence, understanding the social mechanisms leading to the emergence of particular governance arrangements, and how effective they are in addressing climate change and SLR, is limited (Wong et al., 2014; Bisaro and Hinkel, 2016; Oberlack, 2017; Bisaro et al., 2018; Roggero et al., 2018a; Roggero et al., 2018b). An important post-AR5 development has thus been to move beyond descriptions and normative prescriptions about ‘good governance’ to explore which factors help (called enablers) or hinder (called barriers) how social choices are made and implemented on complex issues like climate change and SLR, as elaborated in the next subsection.

<div id="section-4-4-3-2understanding-barriers-to-adaptation-as-governance-challenges"></div>

<span id="understanding-barriers-to-adaptation-as-governance-challenges"></span>
==== 4.4.3.2 Understanding Barriers to Adaptation as Governance Challenges ====

<div id="section-4-4-3-2understanding-barriers-to-adaptation-as-governance-challenges-block-1"></div>

AR5 stated that there are many reasons why adaptation governance is complex (Klein et al., 2014). The first generation of studies that investigated this question empirically identified many (lists of) barriers that people have experienced in adaptation governance in specific case contexts, including political, institutional, social-cognitive, economic, financial, biophysical and technical barriers (Klein et al., 2014). Although insightful for these specific cases, including SLR (Hinkel et al., 2018), accumulation of empirical findings in building theory proved to be limited, and it did not result in more evidence-informed advice to policy makers on how to deal with barriers (Biesbroek et al., 2013; Eisenack et al., 2014).

In response, in a second generation of studies, several frameworks have been proposed and tested to advance scholarship on barriers to adaptation (Eisenack and Stecker, 2012; Barnett et al., 2015; Lehmann et al., 2015; Bisaro and Hinkel, 2016). A frequently used framework was developed by Moser and Ekstrom (2010) who identified and linked key barriers to certain stages of the policy process: understanding, planning and management stages. Moser and Ekstrom (2010) argue that conditions, such as the scope and scale of adaptation, have significant implications for which barriers are activated in the policy process, and how persistent and difficult they are to overcome. This and other frameworks have been applied in a diversity of contexts, providing valuable insights about the governance challenges involved in adapting to climate change and suggestions for improvement (Ekstrom and Moser, 2014; Rosendo et al., 2018; Thaler et al., 2019).

A recent generation of studies takes more theory based approaches and includes contextual factors to analyse the key social mechanisms that explain why adaptation processes are often complex, result in deadlocks, delays or even failure (Biesbroek et al., 2014; Eisenack et al., 2014; Wellstead et al., 2014; Biesbroek et al., 2015; Bisaro and Hinkel, 2016; Oberlack and Eisenack, 2018; Sieber et al., 2018; Wellstead et al.). Such insights are critical as they can be used by practitioners for policy design (e.g., to prevent certain deadlocks from emerging by (re)designing contextual conditions), or provide insights on strategic interventions in ongoing processes to revitalise deadlocked adaptation governance (Biesbroek et al., 2017).

<div id="section-4-4-3-3governance-challenges-in-the-face-of-sea-level-rise"></div>

<span id="governance-challenges-in-the-face-of-sea-level-rise"></span>
==== 4.4.3.3 Governance Challenges in the Face of Sea Level Rise ====

<div id="section-4-4-3-3governance-challenges-in-the-face-of-sea-level-rise-block-1"></div>

There is a wide diversity of governance challenges and opportunities for tackling SLR, with marked differences within and between coastal communities in developed and developing counties. Five salient overarching governance challenges that arise due to distinctive features of SLR were highlighted. This typology is then used to assess how planning, participation and conflict resolution (Section 4.4.4.2), decision analysis methods (Section 4.4.4.3), and enabling conditions (Section 4.5) can help to address these five challenges.

'''Time horizon and uncertainty''' '':'' The long-term commitment to SLR (Section 4.2.3.5) and the large and deep uncertainty about the magnitude and timing of SLR beyond 2050 (Section 4.4.4.3.2), challenge standard planning and decision making practises for several reasons ( ''high confidence'' ; Peters et al., 2017; Pot et al., 2018; Hall et al., 2019; Hinkel et al., 2019). The time horizon of SLR extends beyond usual political, electoral and budget cycles. Furthermore, many planning and decision making practices strive for predictability and certainty, which is at odds with the dynamic risk and deep uncertainty characterising SLR (Hall et al., 2019). Tensions can arise between established risk-based planning that seeks to measure risk, and adaptation responses that embrace uncertainty and complexity (Kuklicke and Demeritt, 2016; Carlsson Kanyama et al., 2019). For example, tensions arise because of the mismatch between the relative inflexibility of existing law and institutions and the evolving nature of SLR risk and impacts (Cosens et al., 2017; Craig et al., 2017; DeCaro et al., 2017). Possible limits of in situ responses to ongoing SLR (e.g., protection and accommodation), bring into question prevailing legal approaches to property rights and land use regulation (Byrne, 2012). In addition, because uncertainty about SLR makes it difficult to decide when to wait and when to act, public actors fear being held accountable for misjudgments (Kuklicke and Demeritt, 2016). The long time horizon and uncertainty of SLR make it difficult to mobilise political will and the leadership required to take visionary action (Cuevas et al., 2016; Gibbs, 2016; Yusuf et al., 2016; Yusuf et al., 2018b).

'''Cross-scale and cross-domain coordination''' : SLR creates new coordination problems across jurisdictional levels and domains, because impacts cut across scales, sectors and policy domains and responding often exceeds the capacities of local governments and communities ( ''medium confidence'' ; Araos et al., 2017; Termeer et al., 2017; Pinto et al., 2018; Clar, 2019; Clar and Steurer, 2019; Sections 4.3.2 and 4.4.2). Local responses are generally nested within a hierarchy of local, regional, national and international governance arrangements and cut across sectors (Cuevas, 2018; Chhetri et al., 2019; Clar, 2019). Furthermore responding to SLR is only one administrative priority amongst many, and the choice of SLR response is influenced by multiple co-existing functional responsibilities and perspectives (e.g., planning, emergency management, asset management and community development) that compete for legitimacy, further complicating the coordination challenge (Klein et al., 2016; Vij et al., 2017; Jones et al., 2019).

'''Equity and social vulnerability:''' SLR and responses may affect communities and society in ways that are not evenly distributed, which can compound vulnerability and inequity, and undermine societal aspirations, such as achieving SDGs ( ''high confidence'' ; Section 4.3.3.2; Eriksen et al., 2015; Foerster et al., 2015; Sovacool et al., 2015; Clark et al., 2016; Gorddard et al., 2016; Adger et al., 2017; Holland, 2017; Dolšak and Prakash, 2018; Lidström, 2018; Matin et al., 2018; Paprocki and Huq, 2018; Sovacool, 2018; Warner et al., 2018a). Costs and benefits of action and inaction are distributed unevenly, with some coastal nations, particularly small island states, being confronted with adaptation costs amounting to several percent of GDP in the 21st century (Section 4.3.3.2). Land use planning for climate adaptation can exacerbate sociospatial inequalities at the local level, as illustrated in a study of eight cities, namely Boston (USA), New Orleans (USA), Medellin (Colombia), Santiago (Chile), Metro Manila (Philippines), Jakarta (Indonesia), Surat (India), and Dhaka (Bangladesh; Anguelovski et al., 2016). Private responses may also exacerbate inequalities as, for example, in Miami, USA, where purchase of homes in areas at higher elevation has increased property prices displacing poorer communities from these areas (Keenan et al., 2018). In Bangladesh, some adaptation practices have enabled land capture by elites, public servants, the military and roving gangs, and resulted in various forms of marginalisation that compound vulnerability and risk (Sovacool, 2018); a reality also faced by many other coastal communities around the world (Sovacool et al., 2015).

'''Social conflict:'''  Ongoing SLR could become a catalyst for possibly intractable social conflict by impacting human activities, infrastructure and development along low-lying shorelines ( ''high confidence'' ). Social conflict refers here to the non-violent struggle between groups, organisations and communities over values, interests, resources, and influence or power, whereby parties seek to achieve their own goals, and may seek to prevent others from realising their goals and possibly harm rivals (Coser, 1967; Oberschall, 1978; Pruitt et al., 2003). SLR impacts that could contribute to conflict include: disruptions to critical infrastructure, cultural ties to the coast, livelihoods, coastal economies, public health, well-being, security, identity and the sovereignty of some low-lying island nations (Sections 4.3.2.4, 4.3.3.2, 4.3.3.6; Mills et al., 2016; Yusuf et al., 2016; Nursey-Bray, 2017; Hinkel et al., 2018). SLR responses inevitably raise difficult trade-offs between private and public interests, short- and long-term concerns, and security and conservation goals, which are difficult to reconcile due to divergent problem framing, interests, values and ethical positions (Eriksen et al., 2015; Foerster et al., 2015; Mills et al., 2016; Termeer et al., 2017; Sovacool, 2018). To some countries, SLR presents a security risk due to the scale of potential displacement and migration of people (Section 4.4.2.6). Climate change, and rising seas in particular, could compound sociopolitical stressors (Sovacool et al., 2015), challenge the efficacy of prevailing legal processes (Byrne, 2012; Busch, 2018; Setzer and Vanhala, 2019), and spark or escalate conflict (Lusthaus, 2010; Nursey-Bray, 2017).

'''Complexity:''' SLR introduces novel and complex problems that are difficult to understand and address ( ''high confidence'' ; Moser et al., 2012; Alford and Head, 2017; Wright and Nichols, 2018; Hall et al., 2019). As a result of the preceding features of the SLR problem, and the complexity of the nonlinear interactions between biogeophysical and human systems, SLR challenges may be difficult to frame, understand and respond to. Often, disciplinary science is not sufficient for understanding complex problems like SLR and traditional technical problem solving may not be well suited for crafting enduring SLR responses (Lawrence et al., 2015; Termeer et al., 2015). SLR poses a challenge for bridging gaps between science, policy and practice (Hall et al., 2019). The complexity and rapid pace of SLR in some localities is also challenging traditional community decision making practices, for example, in some Pacific Island communities (Nunn et al., 2014).

<span id="planning-engagement-and-decision-tools-for-choosing-responses"></span>
=== 4.4.4 Planning, Engagement and Decision Tools for Choosing Responses ===

<div id="section-4-4-4-1introduction"></div>

<span id="introduction-3"></span>
==== 4.4.4.1 Introduction ====

<div id="section-4-4-4-2planning-public-participation-and-conflict-resolution-in-the-face-of-slr"></div>

<span id="planning-public-participation-and-conflict-resolution-in-the-face-of-slr"></span>
==== 4.4.4.2 Planning, Public Participation and Conflict Resolution in the Face of SLR ====

<div id="section-4-4-4-2planning-public-participation-and-conflict-resolution-in-the-face-of-slr-block-1"></div>

'''Land use or spatial planning''' has the potential to help communities prepare for the future and decide how to manage coastal activities and land use taking into account the uncertainty, complexity and contestation that characterise SLR ( ''high confidence'' ; Hurlimann and March, 2012; Hurlimann et al., 2014; Berke and Stevens, 2016; King et al., 2016; Reiblich et al., 2017) Planners work with governing authorities, the private sector, and local communities to integrate and apply tailor-made decision analysis, public participation and conflict resolution approaches that can be institutionalised in statutory provisions, and aligned with informal institutional structures and processes carried out at various scales (Hurlimann and March, 2012; Smith and Glavovic, 2014; Berke and Stevens, 2016).

Planning can play an important role in crafting SLR responses, addressing several of the governance challenges identified above (Section 4.4.3). Planning is future focused and can assist communities to develop and pursue a shared vision, and understand and address SLR concerns in locality-specific ways (Hurlimann and March, 2012; Berke and Stevens, 2016). Planning can help articulate and clarify roles and responsibilities through statutory planning provisions, complemented by non-statutory processes (Vella et al., 2016). It can build social and administrative networks that mobilise cross-scale SLR responses, and facilitate integration of diverse mitigation and adaptation goals alongside other public aspirations and policy imperatives (Hurlimann and March, 2012; Vella et al., 2016). Planning can also facilitate the establishment of collaborative regional forums that cross jurisdictional boundaries and assist local governments and other stakeholders to pool resources and coordinate roles and responsibilities across multiple governance levels, such as the Southeast Florida Regional Climate Change Compact, USA (Shi et al., 2015; Vella et al., 2016). Regulatory planning can be used by governing authorities to steer future infrastructure, housing, industry and related development away from areas exposed to SLR (Hurlimann and March, 2012; Hurlimann et al., 2014; Smith and Glavovic, 2014; Berke and Stevens, 2016).

The extent to which planning is effective in reducing coastal risk, however, varies widely between and within coastal nations (Glavovic and Smith, 2014; Shi et al., 2015; Cuevas et al., 2016; King et al., 2016; Woodruff and Stults, 2016). Planning can fail to prevent development in at-risk locations, and may even accelerate such development, as experienced in settings as diverse as Java, Indonesia (Suroso and Firman, 2018), the Philippines (Cuevas, 2018), Australia (Hurlimann et al., 2014), and the USA (Vella et al., 2016; Woodruff and Stults, 2016). Planning has exacerbated sociospatial inequalities in cities like Boston, USA, Santiago, Chile, and Jakarta, Indonesia (Anguelovski et al., 2016). A study of vulnerability dynamics in Houston, New Orleans and Tampa, USA shows that vulnerability can be reinforced or ameliorated through adaptation planning and decision making processes (Kashem et al., 2016). Regulatory planning may be non-existent in some settings, such as informal settlements, or when used can paradoxically entrench vulnerability and compound risk (Berquist et al., 2015; Amoako, 2016; Ziervogel et al., 2016b). Planning practice is thus both a contributor to and an outcome of local politics and power. Recognising and navigating these challenges is key to realising the promise of planning for reducing SLR risk, and participatory planning processes that reconcile divergent interests are central to this endeavour (Forester, 2006; Smith and Glavovic, 2014; Anguelovski et al., 2016; Cuevas et al., 2016).

'''Public participation''' refers to directly involving citizens in decision making processes rather than only indirectly via voting. Citizen participation is commonplace in public decision making that addresses important societal concerns like SLR (Sarzynski, 2015; Berke and Stevens, 2016; Gorddard et al., 2016; Baker and Chapin III, 2018; Yusuf et al., 2018b). Practices sit along a continuum from manipulation to minimal involvement and more empowering and self-determining practices (Arnstein, 1969; International Association for Public Participation, 2018). Public participation draws on a wide variety of tailored engagement processes and practices, from ‘serious games’ (Wu and Lee, 2015) to role-play simulations (Rumore et al., 2016), and deliberative-analytical engagement (Webler et al., 2016).

There has been a proliferation of public engagement approaches and practices applied to adaptation in recent decades (Webler et al., 2016; Kirshen et al., 2018; Mehring et al., 2018; Nkoana et al., 2018; Yusuf et al., 2018a; Uittenbroek et al., 2019). Increasing citizen participation in adaptation and other public decision making processes shifts the role of government from a chiefly steering and regulating role towards more responsive and enabling roles, sometimes referred to as co-design, co-production, and co-delivery of adaptation responses (Ziervogel et al., 2016a; Mees et al., 2019). Engagement strategies grounded in community deliberation can help to improve understanding about SLR and response options, reducing the polarising effect of alternative political allegiances and worldviews (Akerlof et al., 2016; Uittenbroek et al., 2019). Public participation has also the potential to successfully include vulnerable groups in multi-level adaptation processes (Kirshen et al., 2018), promote justice and enable transformative change (Broto et al., 2015; Schlosberg et al., 2017).

It is widely recognised that authentic and meaningful public participation is important and can help in crafting effective and enduring adaptation responses, but is invariably difficult to achieve in practice (Barton et al., 2015; Cloutier et al., 2015; Sarzynski, 2015; Serrao-Neumann et al., 2015; Berke and Stevens, 2016; Chu et al., 2016; Schlosberg et al., 2017; Baker and Chapin III, 2018; Kirshen et al., 2018; Lawrence et al., 2018; Mehring et al., 2018; Lawrence et al., 2019; Uittenbroek et al., 2019). There is limited empirical evidence that public participation per se improves environmental outcomes (Callahan, 2007; Reed, 2008; Newig and Fritsch, 2009). Major factors determining outcomes are tacit, including trust, environmental preferences, power relationships and the true motivations of sponsor and participants (Reed, 2008; Newig and Fritsch, 2009). Difficulties in realising the anticipated benefits of public participation have been shown in coastal settings including Queensland, Australia (Burton and Mustelin, 2013), Germany’s Baltic Sea (Schernewski et al., 2018), England (Mehring et al., 2018), Sweden (Brink and Wamsler, 2019), and South Africa (Ziervogel, 2019). Research by Uittenbroek et al. (2019) in the Netherlands, for example, shows that public participation objectives are more probable if participation objectives and process design principles and practices are co-produced by community and government stakeholders. In some cities in the Global South, experience shows that a focus on building effective multi-sector governance institutions can facilitate ongoing public involvement in adaptation planning and implementation, and enhance long-term adaptation prospects (Chu, 2016b).

'''Conflict resolution''' refers to formal and informal processes that enable parties to create peaceful solutions for their disputes (Bercovitch et al., 2008). They range from litigation and adjudication to more collaborative processes based on facilitation, mediation and negotiation (Susskind et al., 1999; Bercovitch et al., 2008). Such processes can be used in the public domain to make difficult social choices. Whilst it may be impossible to eliminate controversy and disputes due to SLR, conflict resolution can be foundational for achieving effective, fair and just outcomes for coastal communities (Susskind et al., 2015; Nursey-Bray, 2017). Whereas some responses to social conflict (see definition in Section 4.4.3.3) can be destructive (e.g., resorting to violence), constructive approaches to conflict resolution (e.g., negotiation and mediation) can help disputants satisfy their interests and even have transformational adaptation potential (Laws et al., 2014; Nursey-Bray, 2017). Laws et al. (2014), for example, use the term ‘hot adaptation’ to describe adaptation efforts that harness the energy and engagement that conflict provokes; and create opportunities for public deliberation and social learning about complex problems like SLR. Such an approach has particular relevance in settings most at risk to SLR. Realising this potential is, however, challenging in the face of local politics and the differential power and influence of disputants. These realities have been accounted for in public conflict resolution scholarship and practice for many decades (Forester, 1987; Dukes, 1993; Forester, 2006), and lessons learned are beginning to be applied to adaptation (Laws et al., 2014; Nursey-Bray, 2017; Sultana and Thompson, 2017) and SLR response planning (Susskind et al., 2015). Conflict was turned into cooperation in some villages in floodplains in Bangladesh, for example, by facilitated dialogue and incentivised cooperation between local communities and government, with external facilitator assistance, leading to improved water security in a climate stressed environment (Sultana and Thompson, 2017). At a larger scale, the Mekong River Commission, with its water diplomacy framework, provides an institutional structure and processes, with technical support, and legal and strategic mechanisms, that help to negotiate solutions for complex delta problems and, in so doing, help avert widespread destruction of livelihoods and conflict (Kittikhoun and Staubli, 2018).

Many of the techniques used in planning, public participation and conflict resolution, at times together with decision analysis and support tools, are being applied in combination. In New Zealand, for example, a participatory approach was used to combine dynamic adaptive pathways planning with multi-criteria and real options analysis (Section 4.4.4.3.4) to develop a 100-year strategy to manage coastal hazard risk (Lawrence et al., 2019; see Box 4.1). Public participation thereby helped to shift communities towards a longer-term view and towards considering a wider range of adaptation options and pathways. Such combined approaches are also sometimes referred to as Community Based Adaptation, which involve local people directly in understanding and addressing the climate change risks they face (Box 4.4). These processes and practices are used in many settings, from small, isolated indigenous communities to large-scale coastal infrastructure projects in both the Global North and South. See Table 4.9 in Section 4.4.5 for illustrative examples.

<div id="section-4-4-4-3decision-analysis-methods"></div>

<span id="decision-analysis-methods"></span>
==== 4.4.4.3 Decision Analysis Methods ====

<div id="section-4-4-4-3decision-analysis-methods-block-1"></div>

<span id="introduction-4"></span>
===== 4.4.4.3.1 Introduction =====

Decision analysis methods are formal methods that help to identify alternatives that perform best or well with regard to given objectives. An alternative (also called response option or, as a sequence of options over time: adaptation pathway) is a specific combination of SLR responses (See Section 4.4.3). Each alternative is characterised for each possible future state-of-the world (e.g., levels of SLR or socioeconomic development) by one or several attributes, which may measure any relevant social, ecological, or economic effect associated with choosing and implementing the alternative (Kleindorfer et al., 1993 <sup>[[#fn:r2137|2137]]</sup> ). Attributes commonly used include cost of adaptation alternatives, monetary and non-monetary benefits of the SLR impacts avoided, or net present value (NPV), which is the difference between discounted monetised benefits over time and discounted costs over time. Formal decision analysis is one way to support social choices that is generally suggested for decision support if decisions are complex and involve large investments, as is frequently the case in coastal contexts in the face of SLR.

In order to be effective, decision analysis needs to be embedded in a governance process that accounts for societal needs and objectives (Sections 4.4.4.2 and 4.4.5). This is because decision analysis entails a number of normative choices about the objectives chosen, the criteria used, the specific methods and data applied, the set of alternatives considered, and the attributes used to characterise alternatives. These choices need to reflect the diversity of values, preference and goals of all stakeholders involved in and affected by a decision. Furthermore, decision analysis needs to consider all available knowledge, including all major uncertainties in both climate and non-climate factors, ambiguities in expert opinions, and differences in approaches, because a partial consideration of uncertainty and ambiguity could misguide the choice of adaptation alternatives ( ''high confidence'' ; Renn, 2008 <sup>[[#fn:r2138|2138]]</sup> ; Jones et al., 2014 <sup>[[#fn:r2139|2139]]</sup> ; Hinkel and Bisaro, 2016 <sup>[[#fn:r2140|2140]]</sup> ).

Since AR5, the literature on coastal decision analysis has advanced significantly, specifically addressing the large uncertainty about post-2050 SLR through i) using robust decision approaches instead of expected utility, ii) iterating or adapting decisions over time, and iii) increasing flexibility of responses. Each advance is elaborated below. Furthermore, the coastal decision analysis literature also stresses the consideration of multiple criteria or attributes, because adaptation often involves stakeholders with differing objectives and ways of valuing alternatives (Oddo et al., 2017 <sup>[[#fn:r2141|2141]]</sup> ). Many decision making methods combine each of the three advances highlighted here (Marchau et al., 2019 <sup>[[#fn:r2142|2142]]</sup> ). The suitability of each method depends strongly on the specific context, including available resources, technical capabilities, policy objectives, stakeholder preferences and available information.

<div id="section-4-4-4-3decision-analysis-methods-block-2"></div>

<span id="using-robustness-criteria-instead-of-expected-utility"></span>
===== 4.4.4.3.2 Using robustness criteria instead of expected utility =====

A growing literature on decision analysis of coastal adaptation advocates the use of RDM approaches instead of maximising expected utility approaches (Hallegatte et al., 2012 <sup>[[#fn:r2143|2143]]</sup> ; Haasnoot et al., 2013 <sup>[[#fn:r2144|2144]]</sup> ; Lempert et al., 2013 <sup>[[#fn:r2145|2145]]</sup> ; Wong et al., 2017 <sup>[[#fn:r2146|2146]]</sup> ). The core criterion to be considered for choosing between the two types of approaches is whether one is confronted with a situation of shallow or deep uncertainty ''(high confidence)'' (Lempert and Schlesinger, 2001 <sup>[[#fn:r2147|2147]]</sup> ; Kwakkel et al., 2010 <sup>[[#fn:r2148|2148]]</sup> ; Kwakkel et al., 2016b <sup>[[#fn:r2149|2149]]</sup> ; Hinkel et al., 2019 <sup>[[#fn:r2150|2150]]</sup> ). Uncertainty is shallow when a single unambiguous objective or subjective probability distribution can be attached to states-of-the-world. Uncertainty is deep, when this is not possible, either because there is no unambiguous method for deriving objective probabilities or the subjective probability judgements of parties involved differ (Cross-Chapter Box 4 in Chapter 1; Type 2).

Expected utility approaches can only be applied for identifying an optimal response in situations of shallow uncertainty. This is because these approaches require a probability distribution over states of the world in order to identify the optimal alternative which leads to the highest expected utility (i.e., the probability weighted sum of the utilities of all outcomes under a given alternative and all states-of-the-world; Simpson et al., 2016 <sup>[[#fn:r2151|2151]]</sup> ). A prominent example of this approach is cost-benefit analysis under risk, which assesses expected outcomes across states of the world in terms of NPV (the discounted stream of net benefits). Cost-benefit analysis has several well-known limitations, such as its sensitivity to discount rates and the difficulty to monetise ecological, cultural and other intangible benefits (Section 1.1.4) that have been widely discussed in the climate change literature (Chambwera et al., 2014 <sup>[[#fn:r2152|2152]]</sup> ; Kunreuther et al., 2014 <sup>[[#fn:r2153|2153]]</sup> ; Dennig, 2018 <sup>[[#fn:r2154|2154]]</sup> ).

In the context of coastal adaptation, uncertainty is only shallow if projected SLR does not significantly differ between low end (e.g., RCP2.6) and high end (e.g., RCP8.5) scenarios (Hinkel et al., 2019 <sup>[[#fn:r2155|2155]]</sup> ). The point in time when this is the case (i.e., time of scenario divergence) depends on what difference in expected utility matters to the particular stakeholders involved in a decision. The time of scenario divergence also differs across locations. In locations where the internal sea level variability is large as compared to relative SLR, it takes longer before the differences in sea levels under low end and high end scenarios become apparent. Figure 4.15 illustrates this effect for the ESL projections of this report (Sections 4.2.3.2 and 4.2.3.4), following the approach of Hinkel et al. (2019). Under the assumption that a 10% statistical distance between the distributions of RCP2.6 and RCP8.5 is decision relevant, scenario divergence occurs before 2050 for approximately two thirds of coastal sites with sufficient observational data, but for 7% of locations this occurs later than 2070.

In principle, a single unambiguous probability distribution on future sea levels could also be attained beyond the time of scenario divergence by attributing subjective probabilities to emission scenarios, but individuals may significantly disagree in their subjective probabilities, which again results in deep uncertainty (Lempert and Schlesinger, 2001 <sup>[[#fn:r2156|2156]]</sup> ; Stirling, 2010 <sup>[[#fn:r2157|2157]]</sup> ). For this reason, very few studies that assign subjective probabilities to emission scenarios are found in the literature (Woodward et al., 2014 <sup>[[#fn:r2158|2158]]</sup> ; Abadie, 2018 <sup>[[#fn:r2159|2159]]</sup> ). But even before the year of scenario divergence is reached, uncertainty about relative SLR can be deep, because of deep uncertainties about non-climatic contributors to relative sea level change, such as VLM during or after earthquakes and human-induced subsidence (Cross-Chapter 5 in Chapter 1; Section 4.2.2.4; Hinkel et al., 2019 <sup>[[#fn:r2160|2160]]</sup> ).

Under situations of deep uncertainty, RDM approaches aim to identify alternatives that perform reasonably well (i.e., ‘are robust’) under a wide range of states-of-the-world or scenarios and hence do not require probability assessments. These approaches include minimax or minimax regret (Savage, 1951), info gap theory (Ben-Haim, 2006 <sup>[[#fn:r2162|2162]]</sup> ), robust optimisation (Ben-Tal et al., 2009 <sup>[[#fn:r2163|2163]]</sup> ) and exploratory modelling methods that create a large ensemble of plausible future scenarios for each alternative, and then use search and visualisation techniques to extract robust alternatives (Lempert and Schlesinger, 2000 <sup>[[#fn:r2164|2164]]</sup> ). SLR examples of RDM include Brekelmans (2012) who minimise the average and maximum regret across a range of SLR scenarios for investments in dike rings in the Netherlands and Lempert et al. (2013) <sup>[[#fn:r2165|2165]]</sup> who apply RDM in Hoh-Chi-Minh City '''''.'''''

But even if SLR uncertainty is shallow, RDM are more suitable than expected utility approaches if parties involved or affected by a decision have a low uncertainty tolerance, because the goal of the uncertainty intolerant decision maker is to avoid major damages under most or all circumstances (Hinkel et al., 2019 <sup>[[#fn:r2166|2166]]</sup> ).

An adaptation strategy developed based on the maximisation of expected utility may not meet this goal, because worst case damages occurring can exceed expected damages by orders of magnitude.

The uncertainty tolerance of stakeholders is also determining how large of a SLR range needs to be considered in RDM. Stakeholders (i.e., those deciding and those affected by a decision) that have a high uncertainty tolerance (e.g., those planning for investments that can be very easily adapted) can use the combined ''likely'' range of RCP2.6 and RCP8.5 (0.29–1.10 m by 2100) for long-term adaptation decisions. For stakeholders with a low uncertainty tolerance (e.g. those planning for coastal safety in cities and long term investment in critical infrastructure) it is meaningful to also consider SLR above this range, because a 17% chance of GMSL exceeding this range under RCP8.5 is too high to be tolerated from this point of view (Ranger et al., 2013 <sup>[[#fn:r2167|2167]]</sup> ; Hinkel et al., 2015 <sup>[[#fn:r2168|2168]]</sup> ; Hinkel et al., 2019 <sup>[[#fn:r2169|2169]]</sup> ).

Independent of the debate about whether to apply expected utility or robust decision making approaches, there is an extensive literature that applies scenario-based cost-benefit analysis. For example, this approach has been applied for setting the safety standards of Dutch dike rings (Kind, 2014 <sup>[[#fn:r2170|2170]]</sup> ; Eijgenraam et al., 2016 <sup>[[#fn:r2171|2171]]</sup> ), exploring future protection alternatives for New York (Aerts et al., 2014 <sup>[[#fn:r2172|2172]]</sup> ), Ho Chi Minh City (Scussolini et al., 2017 <sup>[[#fn:r2173|2173]]</sup> ), and for many other locations. Scenario-based cost-benefit analysis differs from cost-benefit analysis under risk discussed above in that scenario-based cost-benefit analysis is not applied to rank alternatives across scenarios, but a ‘separate’ cost-benefit analysis is applied within each emission or SLR scenario considered. While this identifies the optimal alternative under each scenario, it does not formally address the problem faced by a coastal decision maker, namely to decide across scenarios (Lincke and Hinkel, 2018 <sup>[[#fn:r2174|2174]]</sup> ). Nevertheless, the results of scenario-based cost-benefit analysis (i.e., NPV of each alternative under each scenario) provide guidance for decision makers and can also be used as inputs (i.e., as attributes) to robust and flexible decision making approaches

<span id="figure-4.15"></span>
<!-- START IMG -->
<!-- IMG TITLE -->
'''Figure 4.15'''

<span id="figure-4.15-year-of-scenario-divergence-between-extreme-sea-level-projections-for-representative-concentration-pathway-rcp2.6-and-rcp8.5-for-all-tide-gauge-locations-with-sufficient-observational-data-relative-to-a-19862005-baseline-bottom-panel.-time-of-divergence-is-defined-using-a-10-threshold-in-the-statistical-distance-between-the-two-distributions-which-can-be-graphically"></span>
<!-- IMG CAPTION -->
'''Figure 4.15 | Year of scenario divergence between extreme sea level projections for Representative Concentration Pathway (RCP)2.6 and RCP8.5 for all tide-gauge locations with sufficient observational data relative to a 1986–2005 baseline (bottom panel). Time of divergence is defined using a 10% threshold in the statistical distance between the two distributions, which can be graphically […]'''
<!-- IMG FILE -->
[[File:94efe17ef80af59b16ef6c6a70f56b0c IPCC-SROCC-CH_4_15-3000x2996.jpg]]

Figure 4.15 | Year of scenario divergence between extreme sea level projections for Representative Concentration Pathway (RCP)2.6 and RCP8.5 for all tide-gauge locations with sufficient observational data relative to a 1986–2005 baseline (bottom panel). Time of divergence is defined using a 10% threshold in the statistical distance between the two distributions, which can be graphically interpreted as the first year in which at least 10% of the area under the probability distribution function (PDF) of RCP8.5 lies outside of the area under the upper half (i.e., above the 50th percentile) of the PDF of RCP2.6. Upper panels indicate the median and 5–95% range of future extreme sea level (ESL) relative to the 1986–2005 baseline for three tide gauge locations with low variability (Papeete), medium variability (New York) and high variability (Cuxhaven). Locations with low variability have a relatively early scenario divergence.

<!-- END IMG -->
<div id="section-4-4-4-3decision-analysis-methods-block-3"></div>

<span id="adapting-decisions-over-time"></span>
===== 4.4.4.3.3 Adapting decisions over time =====

Irrespective of whether expected utility or robustness criteria are applied, there is ''high confidence'' that an effective way of dealing with large uncertainties is adaptive decision making (also called iterative decision making, adaptive planning or adaptive management), which maintains that decision and decision analysis should be conducted within an iterative policy cycle. This approach includes monitoring of sea level variables and evaluation of alternatives in this light in order to learn from past decisions and collect information to inform future decisions (Haasnoot et al., 2013 <sup>[[#fn:r2175|2175]]</sup> ; Barnett et al., 2014 <sup>[[#fn:r2176|2176]]</sup> ; Burch et al., 2014 <sup>[[#fn:r2177|2177]]</sup> ; Jones et al., 2014 <sup>[[#fn:r2178|2178]]</sup> ; Wise et al., 2014 <sup>[[#fn:r2179|2179]]</sup> ; Kelly, 2015 <sup>[[#fn:r2180|2180]]</sup> ; Lawrence and Haasnoot, 2017 <sup>[[#fn:r2181|2181]]</sup> ). Such a staged approach is especially suitable for coastal adaptation due to the long lead and lifetimes of many coastal adaptation measures and the deep uncertainties in future sea levels (Hallegatte, 2009 <sup>[[#fn:r2182|2182]]</sup> ; Kelly, 2015 <sup>[[#fn:r2183|2183]]</sup> ). Prominent representatives of methods that entail this idea are Dynamic Adaptive Policy Pathways (Haasnoot et al., 2013 <sup>[[#fn:r2184|2184]]</sup> ) and Dynamic Adaptation Planning (Walker et al., 2001 <sup>[[#fn:r2185|2185]]</sup> ). An important prerequisite for any adaptive decision-making approach is a monitoring system that can detect sea level signals sufficiently early to enable the required responses (Hermans et al., 2017 <sup>[[#fn:r2186|2186]]</sup> ; Haasnoot et al., 2018 <sup>[[#fn:r2187|2187]]</sup> ; Stephens et al., 2018 <sup>[[#fn:r2188|2188]]</sup> ).

In recent years, many different frameworks for adaptive decision making have been put forward, including Adaptive Policy Making (Walker et al., 2001 <sup>[[#fn:r2189|2189]]</sup> ), Dynamic Adaptive Policy Pathways (Haasnoot et al., 2013 <sup>[[#fn:r2190|2190]]</sup> ), Dynamic Adaptive Planning (Walker et al., 2013 <sup>[[#fn:r2191|2191]]</sup> ), Iterative risk management (Jones et al., 2014 <sup>[[#fn:r2192|2192]]</sup> ) and Engineering Options Analysis (de Neufville and Smet, 2019). Each frameworks emphasises particular aspects of adaptive decision making and has merits in specific situations depending on the preferences, goals, uncertainties and information at stake (Marchau et al., 2019 <sup>[[#fn:r2193|2193]]</sup> ). Nevertheless, all of these frameworks share the following generic and iterative steps

# Set the stage: Identify current situation, objectives, options (alternatives) and uncertainties.
# Develop a dynamic plan, which consists of a basic plan plus contingency actions to be carried out based on observed triggers.
# Implement basic plan and monitor system for triggers.
# Monitor and act upon triggers.

<div id="section-4-4-4-3decision-analysis-methods-block-4"></div>

<span id="increasing-flexibility-of-responses"></span>
===== 4.4.4.3.4 Increasing flexibility of responses =====

An idea closely related to adaptive decision making is to keep future alternatives open by favouring flexible alternatives over non-flexible ones. An alternative is said to be ‘flexible’ if it allows switching to other alternatives once the implemented alternative is no longer effective. For example, a flexible protection approach would be to build small dikes on foundations designed for higher dikes, in order to be able to raise dikes in the future should SLR necessitate this.

A prominent and straightforward method that addresses the objective of flexibility is adaptation pathways analysis (Haasnoot et al., 2011 <sup>[[#fn:r2194|2194]]</sup> ; Haasnoot et al., 2012 <sup>[[#fn:r2195|2195]]</sup> ), which is one component of Dynamic Adaptive Policy Pathways. The method graphically represents alternative combinations of measures over time together with information on the conditions under which alternatives cease to be effective in meeting agreed objectives, as well as possible alternatives that will then be available. As time and SLR progress, monitoring may trigger a decision to switch to another alternative. Adaptation pathway analysis has been widely applied both in the scientific literature as well as in practical cases. Applications after AR5 include Indonesia (Butler et al., 2014 <sup>[[#fn:r2196|2196]]</sup> ), New York City (Rosenzweig and Solecki, 2014 <sup>[[#fn:r2197|2197]]</sup> ), Singapore (Buurman and Babovic, 2016 <sup>[[#fn:r2198|2198]]</sup> ) and Australia (Lin and Shullman, 2017 <sup>[[#fn:r2199|2199]]</sup> ). In New Zealand, the method has been included in national guidance for coastal hazard and climate change decision making (Lawrence et al., 2018 <sup>[[#fn:r2200|2200]]</sup> ). There is ''high confidence'' that the method is useful in interaction with decision makers and other stakeholders, helping to identify possible alternative sequences of measures over time, avoiding lock-in, and showing decision makers that there are several possible pathways leading to the same desired future (Haasnoot et al., 2012 <sup>[[#fn:r2201|2201]]</sup> ; Haasnoot et al., 2013 <sup>[[#fn:r2202|2202]]</sup> ; Brown et al., 2014 <sup>[[#fn:r2203|2203]]</sup> ; Werners et al., 2015 <sup>[[#fn:r2204|2204]]</sup> ).

Alternatives can also be characterised through multiple attributes such as costs, effectiveness, co-benefits, social acceptability, etc., which in turn can be used in multi-attribute decision making methods (Haasnoot et al., 2013 <sup>[[#fn:r2205|2205]]</sup> ). An important attribute is transfer cost, which is the cost of course correction (switching from one alternative to another), reflecting the potential for path dependency (Haasnoot et al., 2019 <sup>[[#fn:r2206|2206]]</sup> ). Delaying decisions and opting for flexible measures introduces extra costs, such as transfer costs. Also, flexible measures are often more expensive than inflexible ones, and damages may occur whilst delaying the decision. An important question therefore is whether it is cheaper to implement a flexible measure now or to wait and implement a less flexible (i.e., cheaper) measure later in time when more information is at hand.

Technically more demanding methods such as real-options analysis (Dixit et al., 1994 <sup>[[#fn:r2207|2207]]</sup> ), and decision tree analysis (Conrad, 1980 <sup>[[#fn:r2208|2208]]</sup> ), can also find pathways that are economically efficient in terms of flexibility and timing of adaptation. There is little application of these approaches in the SLR literature. For example, Woodward et al. (2014) applied real-options analysis to determine flood defences around the Thames Estuary, London, England; Buurman and Babovic (2016) for climate-proofing drainage networks in Singapore; Dawson et al. (2018) for coastal rail infrastructure in southern England; and Kim et al. (2018) for assessing flood defences in southern England. A requirement for applying real-options analysis and decision tree analysis is to quantify today how much will have been learned at a given point in time in the future. The few applications of these methods to SLR-related decisions in the literature have generally used ad-hoc assumptions. For example, Woodward et al. (2011) assumed either perfect learning (i.e., in 2040, which SLR trajectory is occurring will be known) or no learning (i.e., uncertainty ranges and confidence in these remains as today). Others have derived learning rates from comparing past progress in SLR projections and then applied these to the future. An example is given by Dawson et al. (2018) who derive learning rates from the 2002 and 2009 SLR projections of the UK Climate Impacts Programme and apply these in real-options analysis.

<div id="section-4-4-4-3decision-analysis-methods-block-5"></div>

<span id="research-needs"></span>
===== 4.4.4.3.5 Research needs =====

Four general gaps can be identified in the literature. First, the generation of SLR information is insufficiently coupled to the use of this information in decision analysis. This constitutes a limitation, as different coastal decision contexts require different decision analysis methods, which in turn require different SLR information. Specifically, applications of decision analysis methods generally convert existing sea level information to fit their method, often misinterpreting the information, making arbitrary assumptions or losing essential information in the process (Hinkel et al., 2015 <sup>[[#fn:r2210|2210]]</sup> ; Bakker et al., 2017 <sup>[[#fn:r2211|2211]]</sup> ; Van der Pol and Hinkel, 2018) . Second, with the exception of adaptation pathway analysis, methods of robust and flexible decision making are under-represented in the literature despite their suitability (Van der Pol and Hinkel, 2018) . Third, research is necessary to compare the various methods, to identity which methods are most suitable in which context and to develop consistent categorisations of methods (Hallegatte et al., 2012 <sup>[[#fn:r2212|2212]]</sup> ; Haasnoot et al., 2013 <sup>[[#fn:r2213|2213]]</sup> ; Hinkel et al., 2015 <sup>[[#fn:r2214|2214]]</sup> ; Watkiss et al., 2015 <sup>[[#fn:r2215|2215]]</sup> ; Dittrich et al., 2016 <sup>[[#fn:r2216|2216]]</sup> ; Suckall et al., 2018 <sup>[[#fn:r2217|2217]]</sup> ) . Fourth, future research needs to address how to embed decision analysis better in real world planning and decision making processes, recognising that adaptation to SLR is a multi-stakeholder process often characterised by conflicting interests and interdependence between stakeholders (Section 4.4.3). Addressing these gaps requires closer cooperation between SLR sciences, decision science, and planning and governance scholars. An underlying challenge is to design and integrate relevant formal decision making approaches into the heterogeneous reality of local planning and decision making cultures, institutions, processes and practices, often with community-specific needs and requirements (see Box 4.4).

<div id="section-4-4-4-3decision-analysis-methods-block-6" class="box"></div>

<span id="box-4.4-community-based-experiences-canadian-arctic-and-hawkes-bay-new-zealand"></span>