Targeting adaptation to safeguard sustainable development against climate-change impacts


 The international community has committed to achieve 17 Sustainable Development Goals (SDGs) by 2030 and to enhance climate action under the Paris Agreement. Yet achievement of the SDGs is already threatened by climate-change impacts. Here we show that further adaptation this decade is urgently required to safeguard 68% of SDG targets against acute and chronic threats from climate change. We analyse how the relationship between SDG targets and climate-change impacts is mediated by ecosystems and socio-economic sectors, which provides a framework for targeting adaptation. Adaptation of wetlands, rivers, cropland, construction, water, electricity and housing in the most vulnerable countries should be a global priority to safeguard sustainable development by 2030. We have applied our systems framework at the national scale in Saint Lucia and Ghana, which is helping to align National Adaptation Plans with the SDGs, thus ensuring that adaptation is contributing to, rather than detracting from, sustainable development.

Closing this research gap requires identifying an intermediary between SDG targets and climate-42 change impacts that provides a direct entry-point into more granular and spatially-explicit 43 decision-making 5-7 . Ecosystems and socio-economic sectors can provide such an intermediary, as 44 these are critical both for the SDGs and for climate action. Previous research has highlighted the 45 role of ecosystems and socio-economic sectors for achieving SDG targets 8-14 , which has been 46 used to inform SDG decision-making in practice 15 . Past research has also estimated risks from 47 climate change on ecosystems and socio-economic sectors [16][17][18][19][20][21][22] , which has informed spatially-48 explicit climate action. Studies that integrate SDGs and climate-change impacts in the context of 49 ecosystems and socio-economic sectors has so far focused on climate mitigation 20 . However, no 50 such integration exists in the context of climate adaptation. 51 water, transport, energy or medicines), supporting (habitat), and cultural services (heritage, 78 recreational) 8,25 . Socio-economic sectors and their services are sub-classified into three categories: 79 utilities (electricity, transport, water), primary/secondary (manufacturing, mining, construction) and 80 tertiary (public administration, education, healthcare, amongst others) (see Supplementary Tab 1 81 and 2.1). Taking a service-centric approach, our systems framework is based on two phases, 82 analysing how: 83 i) each of the SDG targets can be influenced by ecosystems and socio-economic 84 sectors; 85 ii) each of these ecosystems and socio-economic sectors can be affected by acute 86 and chronic climate-change impacts (see Supplementary Information Tab 3.1 and 87

for evidence). 88
We integrate i) and ii) to derive how SDG targets are affected by, and can be safeguarded against, 89 climate-change impacts. Our findings fill a critical gap, because we identify that adapted 90 ecosystems can safeguard 105 of all 169 SDG targets (62%); adapted utility infrastructure can 91 safeguard 121 targets (72%); adapted primary/secondary sectors can safeguard 67 targets (40%); 92 and adapted tertiary sectors can safeguard all SDG targets. 93 We then apply our systems framework to quantify how near-term risk from acute and chronic 94 climate-change impacts can threaten SDG targets. Near-term risk is here defined based on the 95 Intergovernmental Panel on Climate Change Fifth Assessment Report (IPCC AR5) 16 as a 96 function of high probability/large magnitude hazard, persistent exposure and vulnerability by the 97 2030s (aligned with the SDG timeline). Finally, we demonstrate how our systems framework is 98 applicable at the national scale to align SDG targets with commitments under the Paris 99 Agreement. 100 SDG targets influenced by ecosystems and socio-economic sectors 101 In the first phase, we identify how each SDG target is influenced by ecosystems and socio-102 economic sectors, differentiating by direct, interdependent, and indirect influences. 103 We define direct SDG influences as cases where SDG targets are described in terms of a sector's 104 service, in line with previous research 11 (see Methods). Collectively, we find that 141 of all 169 105 SDG targets (83%) are directly influenced by ecosystems and socio-economic sectors, i.e. these 106 targets are directly described in terms of these sectors' services. With respect to ecosystems, 40 107 SDG targets (24%) are directly influenced by ecosystem services (Figure 2, coloured shading). In 108 addition to targets under SDG14 ('life below water') and SDG15 ('life on land') that explicitly 109 mention ecosystems 14 , we find that targets under 12 different SDGs are directly influenced by 110 the regulating, provisioning, supporting, and cultural services that ecosystems provide. 111 With respect to socio-economic sectors, 29 SDG targets (17%) are directly influenced by 112 infrastructure services provided by utility sectors, such as electricity or transportation services; 13 113 targets (8%) are directly influenced by services provided by primary/secondary sectors, including 114 manufacturing or construction services; and 122 targets (72%) are directly influenced by services 115 provided by tertiary sectors, such as public administration or healthcare services. 116 We distinguish ecosystems or socio-economic sectors (hereafter referred to as sectors) for 117 practical reasons, acknowledging that multiple sectors act interdependently to provide services. For 118 example, water services can be provided by both rivers and physical utilities; ecosystem services 119 such as flood protection can complement or substitute physical infrastructure services; and 120 socio-economic governance services enable equitable ecosystems management. Additionally, 121 cultural services permeate through and across all sectors 26 . We consider interdependencies in direct 122 SDG influences by accounting for whether SDG target progress requires unique, cross-sectoral, or 123 substitutable contributions. Whilst some SDG targets are described in terms of a single sector's 124 service only (unique influence), other targets are described in terms of multiple sectors' services 125 (cross-sectoral influence) or in terms of a sector's service that can be substituted by a different 126 sector (substitutable influence). 127 We find that 68 SDG targets (40%) are influenced uniquely by a single sector's service (Figure 2, 128 magenta shading), where targets under SDG16 ('peace') require most unique influences. In 129 contrast, 53 targets (31%) are influenced by multiple services from different sectors, where each 130 sector provides a cross-sectoral influence to SDG target progress (Figure 2, blue shading). Targets 131 under SDG11 ('sustainable cities') require most cross-sectoral influences. In addition, 20 targets 132 (12%) are influenced by a sector's service that is substitutable by another sector (Figure 2,

Services of land-uses (sectors)
Ecosystems and socio-economic sectors influenced by climate-change 153 impacts 154 In the second phase, we identify how the quantity and/or quality of sectors' services can be 155 threatened by acute and chronic climate-change impacts (Figure 3, red and blue shading). 156 Comparing the expected change in hazard frequency across all acute climate-change impacts under 157 a 1.5°C and 2°C warming scenario, we find the largest increase for extreme temperatures (Figure  158 3, percentages and arrows). Based on IPCC AR5 classification of highest global near-term risk 16 , the 159 provision of services from six sectors is at high near-term risk from extreme temperatures ( Figure  160 3, exclamation marks). This includes the provision of water, food, electricity, construction and 161 housing services, which is likely to have most devastating consequences for sectors or populations  Grassland, savanna & shrubland Forest (Tro pical & te mpe rate ) Barren (No n-ve ge tate d are as) Housing & real estate Number of sectors with high global near-term risk of climate-change impact

Services of land-uses (sectors)
M M Chronic climate-change impacts predominantly influence ecosystems and socio-economic sectors 180 negatively, with some regional positive effects. For example, chronic warming is projected to 181 reduce agricultural yields globally, but may increase yields in northeast China and the UK 31 . 182 Evidence shows that limiting warming to 1.5°C can substantially reduce the frequency of all 183 chronic climate-change impacts, especially for precipitation, chronic warming, and the drying trend 184 climate-change impacts, which have either negative or regionally positive effects, can threaten 37% 197 more SDG targets than they can support through opportunities. When combining direct and indirect 198 influences, we find that progress for all SDG targets can be affected by acute and/or chronic 199 climate-change impacts on sectors' services. 200 Applying this systems framework globally, we identify that the seven sectors at high near-term risk 201 from extreme temperatures and chronic warming can directly influence 36% of SDG targets ( Figure  202 4, exclamation marks). Especially affected are SDG6 ('clean water'), SDG7 ('energy'), SDG9 203 ('innovation and infrastructure'), SDG11 ('sustainable cities') and SDG12 ('responsible 204  Tailoring adaptation to safeguard SDG progress 214 We show how our systems framework can be used to tailor adaptation, which we then discuss in 215 the context of our global near-term risk findings. 216 In tailoring adaptation to SDG targets, we focus on how sectors influence targets (unique, cross-217 sectoral, substitutable, indirect). Where SDG targets are directly influenced by a single climate-218 sensitive sector, adaptation can focus uniquely on that sector. We find that adaptation of public 219 administration, which includes implementing just policy, can uniquely safeguard most SDG 220 targets (27%). 221 Where SDG targets are influenced by different climate-sensitive sectors, cross-sectoral adaptation is 222 needed. Adapting the public administration, finance & insurance, and forest sector can safeguard 223 most SDG targets through cross-sectoral contributions (17%, 11% and 11% of SDG targets, 224 respectively) where each service provides an independent contribution to target achievement. 225 In cases where SDG targets are influenced by climate-sensitive sectors that provide substitutable 226 functions, decision-makers can choose where to adapt. We find that protecting or enhancing 227 ecosystem services -including rivers & lakes, forest, and wetlands & peatlands -can safeguard 228 10%, 8%, and 8% of SDG targets, respectively, through substituting socio-economic sectors. 229 Awareness of indirect SDG influences is critical for decision-makers to maximise co-benefits of 230 adaptation. Given our finding that SDG8 and SDG5 are influenced by most indirect influences 231 (eight and six, respectively) compared to direct ones, embedding local economic opportunities 232 and gender considerations across all adaptation policies can maximise SDG contributions. 233 In tailoring adaptation to reduce risk, we differentiate by the three components of risk: hazard, 234 exposure, and vulnerability. Firstly, decision-makers might focus adaptation on areas projected to 235 experience proportionally more frequent/severe hazards. Adaptation options could include 236 greening to reduce extreme temperatures in cities or restoring wetlands to reduce flood severity. 237 Secondly, decision-makers might tailor adaptation based on how hazards expose sector's supply 238 ('land/resources', 'physical capital', 'labour') or demand (Supplementary Information Tab 3.2). For 239 example, on agricultural croplands there is evidence that extreme temperatures mainly expose 240 outdoor-working agricultural 'labour', whilst floods mainly affect agricultural 'physical capital' 241 ( Figure 5, horizontal sector comparison). Therefore, whilst working hour policies might reduce 242 exposure on 'labour', better flood protection is needed to reduce exposure on 'physical capital'. 243 Thirdly, decision-makers may tailor adaptation based on which sectors/populations are more 244 vulnerable to hazard exposure. For example, already threatened ecosystems/species are more 245 sensitive to hazards, and poor agricultural workers who are already working under insecure 246 arrangements are less capable to adapt 16 . Vulnerability-based adaptation may therefore focus on 247 integrating biodiverse habitats to connected networks to enable threatened plant/animal species 248 to adapt in response to climate change 32 , or on social protection policies to increase adaptive 249 capacity. 250  contribution of ecosystem regulating and cultural services to the SDGs over time 37 . Therefore, 280 unless these highly productive and non-substitutable ecosystems are urgently protected against 281 near-term risk, it will not be possible to safeguard the SDG targets they influence. 282 A national adaptation roadmap for climate-resilient development these national applications provided evidence, at high spatial granularity, of how and where to 300 prioritise adaptation in the context of the SDGs, thereby informing NAPs and NDC revisions. 301 As more nations around the globe revise their SDG and climate commitments, including in 302 relation to covid-19 recovery packages, we propose a six-step national adaptation roadmap. The 303 roadmap is centered around an iterative stakeholder-led process and a multifunctional landscape 304 approach 39 that consider interdependencies and trade-offs between ecosystems, socio-economic 305 sectors, and the needs of different beneficiaries. It includes the following steps: 306 (1) identify current and desired levels of ecosystem and socio-economic service provision in 307 relation to needs and SDG-aligned targets 15 across spatially vulnerable populations (e.g. 308 using citizen-science data 40 ); 309 (2) combine top-down and bottom-up climate assessments 4 (e.g. spatial risk analyses 41 , 310 statistical methods 42 , or mixed methods) to assess how current and desired future service 311 provision is at-risk from climate-change impacts, using unified metrices where possible; 312 (3) prioritise adaptation needs across sector, areas, and hazards, considering considers interdependencies between sectors, and differentiates by different acute and chronic 359 physical climate-change impacts. We apply this systems framework to high near-term risk from 360 climate-change impacts, as defined by the IPCC Table TS.4 16 as risk resulting from the following 361 criteria: large magnitude, high probability, or irreversibility of impacts; persistent exposure or 362 vulnerability; limited potential to reduce risks through adaptation or mitigation. These risk levels 363 integrate probability and consequence over the widest possible range of potential outcomes 364 resulting from the interaction of climate-related hazards, exposure, and vulnerability, based on 365 available evidence 16 . 366

Research design 367
Our framework is based on best practices for evidence mapping processes in adaptation

387
Step 1: Aim & concept 388 The aim of this research was to identify a framework of how the SDG targets can be influenced 389 by different climate-change impacts in order to provide a direct entry point for decision-makers 390 to prioritise and target adaptation in the context of the SDGs. Previous literature has identified 391 Climate-change impacts Step 1: Aim & concept

SDG targets
Step

Synthesis
Step 2: Data source & selection that this additional step requires a more granular analysis that can be contextualised to suit the 392 specific needs of decision-makers at different scales (global, national, public and private sector, 393 academic modelling), which necessitates a role for an intermediary 4 . Therefore, addressing our 394 aim involved designing a framework to ensure relevance for decision-makers to inform 395 adaptation, via conceptualising an intermediary between SDG targets and climate-change 396 impacts. We identified the following set of criteria for such an intermediary: We base our intermediary on the original land-cover/land-use classification by USGS 44 , which 411 was developed using strict criteria that spatial units are geographically exclusive and exhaustive. 412 Given that USGS was the first classification of land-cover and land-use, a range of global 413 ecosystem and global land cover classifications build on it. Its spatial exhaustion across the globe 414 suggests that this classification allows for a flexible approach for further disaggregation on a 415 national and/or regional scale. In addition, geospatial data (at 250m resolution) for the terrestrial 416 and freshwater domain is available 56 . We updated the categories in line with the SEEA-updated 417 USGS categories (of major ecosystem types) to classify natural/semi-natural ecosystems 56  • If 'Yes', classify as 'substitutable SDG influence'. 456

Definition of interdependent SDG influence (direct SDG influences only): 457
Unique SDG influence. A unique influence is identified when a sector's service 458 provides independent, singular contributions towards achievement of an SDG 459 target. For example, target 16.3 "Promote the rule of law at the national and 460 international levels" is uniquely influenced by (i.e. directly described in terms of 461 only) the 'law enforcement' services provided by the public administration sector. 462 This function cannot be substituted by the services of another sector. 463

Cross-sectoral SDG influence. A cross-sectoral influence is identified when a sector's 464
service provides independent, cross-sectoral contributions towards achievement of 465 an SDG target. For example, target 11.4 "Strengthen efforts to protect and 466 safeguard the world's cultural and natural heritage" requires the 'cultural heritage' 467 services from the arts & recreation sector as well as the 'natural heritage' services 468 from the forest sector. The services from these sectors must both be safeguarded to 469 ensure target achievement. 470 Substitutable SDG influence. A substitutable influence is identified when sectors 471 provide a service that can be substituted by another sector. In such a case, various 472 sectors provide the same service to achieve progress towards the SDG target, 473 presenting decision-makers with a choice of how to safeguard target achievement in 474 the face of acute and chronic climate-change impacts. For example, target 6.1 475 "achieve universal and equitable access to safe and affordable drinking water", can 476 be achieved through the water provision services directly abstracted from 477 mountainous rivers or via water utilities. 478

Definition of climate change influence 515
An influence is identified if published evidence indicates that a climate-change 516 impact affects the quantity or quality of services from the sector via effects on the 517 three supply factors ('land/natural resources', 'physical capital', or 'labour') or on 518 directly quantifiable 'demand'. Thereby, our definition of climate-change influence 519 for sectors focuses on the exposure of sectors' functioning to climate-change 520 impacts. We refer to climate-change impacts as 'climate-related drivers of impact' 521 summarised by the IPCC 16 . If reference is made to 'coastal infrastructure', any socio-522 economic is considered to be potentially affected. If the literature makes reference to 523 'extreme events', the definition of the IPCC 6 was used, and the following climate-524 change impacts were included: extreme precipitation, damaging cyclone, extreme 525 temperature, flooding, storm surge. The definition excludes the ways in which 526 climate change can result in economic market readjustments (e.g. increased prices 527 due to shortage of supply following extreme impacts). 528 Step 2: Data source & selection 529 The search for scientific evidence was conducted in four different stages (in order of search): 1) 530 Tier 1 journals, 2) IPCC assessments (including 'Global Warming of 1.5°C' 20 , 'Fifth Assessment 531 Report' 16 , 'Climate Change and Land' 34 ), other peer-reviewed articles, and pre-prints, and 3) 532 'Grey literature' (reports from international organisations, national and subnational agencies). 533 The search for evidence was performed through Web of Science and Google Scholar, which 534 were chosen given their high speed of inclusion of related articles and the inclusion of preprints. 535 A google search was used to identify evidence from stage 3 ('Grey literature'). English was used 536 for the evidence search, as it is the most employed language and considered as the international 537 academic language 58 . 538 Results were filtered by date and only the most recent publication was included in the evidence 539 mapping, given that there might be many different publications for each influence 540 ( Supplementary Information Tab 3.1 and 3.2). We did not conduct a meta review of the evidence 541 to characterise the quality of the evidence, but hope to mitigate this aspect through our 542 prioritised search in different stages and using the most recent available evidence. 543 The evidence for each influence was screened against the predetermined definitions, incl. 544 inclusion and exclusion criteria, firstly by two of the authors. The influences were then reviewed 545 and enriched by other authors. Finally, ambiguous influences were discussed, and inclusion and 546 exclusion criteria refined through facilitated discussion until a consensus was reached. 547 Step 3: Analysis & presentation 548 The evidence was characterised through a binary process: whether or not there is an influence. 549 For the SDG influences, we described the evidence in terms of absolute numbers and 550 percentages of SDG targets potentially influenced. We categorised sectors into the following 551 categories to provide useful implications for decision-making: ecosystems; economic sectors 552 (primary/secondary, utility) and social sectors (tertiary). We analysed results at the service level to 553 identify where the same service can be derived from different ecosystems or socio-economic 554 sectors. We did not assess the magnitude of service contributions to each target, because such 555 information is not available at the global scale across all ecosystems and socio-economic sectors, 556 and is highly context-specific. For the climate influences, we identified whether there is a 557 negative or positive influence from climate-change impacts on a supply factor ('land/natural 558 resources', 'physical capital', 'labour') or 'demand' for each sector. 559 Finally, we combined the SDG influences (Phase 1) and the climate influences (Phase 2) in order 560 to compute how each SDG target can be affected by climate-change impacts via sectors' 561 services. For example, if there is published evidence of a negative effect of chronic warming on 562 agricultural food production, we compute the number of SDG targets directly and indirectly 563 influenced by cropland-based food production. 564 We further apply findings from (Phase 1) and (Phase 2) and near-term risk to discuss potential 565 adaptation options tailored to SDG targets and to different climate-change impacts. We 566 classified adaptation to reduce risk based on the three factors that make up risk as defined by the 567 IPCC 59 . These include risk as resulting from the interaction of the: a) climate-change hazard -568 the probability and severity of a climate-change impact event; b) exposure -the land or 569 resources, physical capital, workers or demand for a sector's services subject to a climate-change 570 impact; and c) vulnerability -the sensitivity of a sector to impacts and the capacity of a 571 population/ society to deal with a hazardous impact 29 . Our focus lies on planned rather than 572 autonomous adaptation, and anticipatory rather than reactive adaptation. 573 To inform adaptation ( Figure 5), we counted the number of climate-change impacts affecting 574 each supply factor ('land/natural resources', 'physical capital', 'labour') or 'demand' for each 575 sector based on available evidence on how the climate-change impact affects each sector. We 576 averaged this count per category (ecosystem, utilities, primary/secondary, tertiary) and used an 577 equal interval rank for each category and each supply and demand factor based on the count of 578 climate-change impacts into 'low' (less than 4 impacts), 'moderate' (less than 8 impacts); 'high' 579 (more than 8 of the 12 potential climate-change impacts affecting the supply or demand factor 580 for each category). 581 For our global application, we compute the SDG targets that can be affected by, and require 582 safeguarding from, high near-term risk from climate change. To identify global near-term risk, 583 we used IPCC's key sectoral risk ranking (Table TS.4) 16 , the best globally available ranking of risk 584 across climate-change impacts and sectors. IPCC's risk categorisation was developed based on 585 the following specific criteria: large magnitude, high probability, timing or irreversibility of 586 hazard, persistent exposure or vulnerability contributing to risks, and limited potential to reduce 587 risk through adaptation. We marked a sector as being at high global near-term risk if IPCC's 588 Table TS Limitations 597 There are many ways that sectors and services can be categorised. To provide a systems 598 framework that is transferable across nations, we base our classification on an original land-599 cover/land-use classification (differentiating by ecosystems and socio-economic sectors). We 600 categorised the services provided by each sector, acknowledging the difficulties with allocating 601 services to ecosystems 60 . Instead of focusing on sectors and services, one might also focus on 602 systems of receptors, as discussed in the literature 4 . We opted for the internationally-classified 603 set, given our expectation that the framework can be applied with international and national 604 accounting data across environmental and socio-economic sectors which is typically presented in 605 the form of the sector categories applied (see SEEA 56 ). We used a granular scale for the socio-606 economic sectors, as ISIC provides an overview of services linked to each sector. A similarly 607 granular scale for ecosystems and the specific services associated with each ecosystem is not yet 608 available, but is currently being developed by the SEEA. Future work should update the 609 categories accordingly. 610 We focused on the climate-change impacts as defined by the IPCC AR5 report and acknowledge 611 the absence of 'fire' 61 in this list. We did aim to mitigate this aspect by influences those climate-612 change influences whereby fires are exacerbated by droughts. Despite its limitations, we chose 613 IPCC AR5 climate-change impacts as it is the only source of global evidence which includes 614 multiple climate-change impacts and sectors. 615 Additionally, there are many ways that the range of adaptation options that are available could be 616 characterised 62 , so no characterisation is likely to be universally agreed upon. Following previous 617 researchers 63,64 , we acknowledge that adaptation includes (1) recognition activities (activities that 618 demonstrate awareness), (2) groundwork activities (preliminary steps that inform action but do 619 not constitute actual policy changes, such as vulnerability assessments or conceptual tools), and 620 (3) adaptation action (tangible options taken to 'alter institutions, policies, programs, built 621 environments, or mandates in response to experienced or predicted risks of climate change'). 622 Given that the aim of the paper is to target specific adaptation options to inform national public 623 adaptation decisions across sectors, we focused on the third category of adaptation actions with 624 the aim of reducing risks of climate change. 625 This paper is based on evidence mapping of influences. For some influences between SDG 626 targets and sectoral services, or between sectoral services and climate-change impacts there 627 might not be published evidence yet. The absence of identified evidence does not mean the 628 absence of an influence. Nevertheless, a focus on influences as captured in this paper are based 629 on existing published evidence and are therefore replicable and supported. 630 It is possible for existing literature to make erroneous inferences on influences, especially when 631 based on grey literature. Moreover, some climate-change impacts or SDG influences might be 632 under-researched and therefore not identified. We aimed to mitigate this aspect by reviewing 633 several studies for each influence and by discussing any potential issues or ambiguities with the 634 authors of this paper, which span a range of disciplines and topical expertise (including 635 geography, engineering, social science and ecology as well as topics from ecosystems & 636 biodiversity, infrastructure, climate risk analysis, SDG target mapping, adaptation, amongst 637 others). Moreover, it is also possible for sectors to have negative influences on the SDGs. Whilst 638 we do not specifically assess trade-offs, our framework provides an indication where a sector's 639 services can influence SDG targets, which can be used to identify negative influences. Additional 640 research should also assess negative interdependent influences, for example to assess the effect 641 of socio-economic sectors on ecosystems. 642 With respect to our global application, the identification of highest near-term risk is based on the 643 best available evidence of changing hazard frequency ( Supplementary Information Tab 3.3) and 644 global sectoral near-term risk from the IPCC AR5 16 . Whilst this IPCC assessment is based on 645 expert elicitation and only covers selected risk, it's use nevertheless provides a replicable and 646 evidence-based overview of the type of sectoral risks to different climate-change impacts. The 647 framework can be further applied to an updated overview of global sectoral risk from climate-648 change impacts, such as for example IPCC's AR6 or other global studies. 649

Data availability 650
The data that support the findings of this study are available within the paper and its 651 Supplementary Information.     SDG targets in uenced by ecosystems and socio-economic sectors. Each rectangle represents one SDG target; magenta shading denotes a unique direct in uence; blue represents a cross-sectoral direct in uence; and green denotes a substitutable direct in uence for achieving the SDG target. Grey indicates an indirect in uence, where there is published evidence that improving the quality/quantity of the sector's services can help achieve the SDG target. White shading indicates the absence of identi ed evidence.
Evidence is reported in Supplementary Information Tab 3.1.

Figure 3
Ecosystems and socio-economic sectors in uenced by acute and chronic climate-change impacts. Red shading denotes evidence of a negative impact, blue shading highlights a global net-negative impact, but potential positive regional effect of climate-change impact on services. White shading indicates the absence of identi ed evidence. Evidence is reported in Supplementary Information Tab 3.2. Exclamation marks represents high near-term risk based on IPCC AR5 TS.416. Percentages for climate-change impacts signify changes in frequency under a 1.5 and 2°C scenario (see Supplementary Table 2 and Supplementary Information Tab 3.3), whereby the symbol * suggests that no quanti ed evidence was identi ed.

Figure 4
Systems framework, showing (from left to right): Percentage of sectors under each category (ecosystems, utilities, primary/secondary, tertiary) impacted by acute (red Sankey lines) and chronic climate-change impacts (blue Sankey lines); quantity of acute (red bars) and chronic climate-change impacts (blue bars) on sectors; quantity of direct (dark green bars) and indirect SDG target in uences (light green bars) from sectors; and the percentage of SDG targets under each goal directly in uenced (green Sankey lines) by each sector category. Exclamation marks (left) denote high global near-term risk16.

Figure 5
Tailoring adaptation to climate-change impacts. Example adaptation options applicable to acute and chronic climate-change impacts, based on average count of evidence on how climate-change impacts affect ecosystems and socio-economic sectors. Absence of colour indicates the absence of identi ed evidence. a) Effects of acute and chronic climate-change impacts on supply of and demand for services of different sector categories; b) Ability of adaptation options to safeguard service provision in the face of climate-change impacts. The example list of adaptation options is not extensive.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. CCadaptationSDGssupplementaryinformationFuldauer2021.xlsx