Hazard indicators are projected for two CC scenarios RCP 2.6 (low emissions scenario) and 8.5 (High emissions scenario) and two periods (2046-2065) and (2081-2100). The study addresses the problem of an evident lack of high-resolution wave simulations comprising the Atlantic islands, by the expansion of the Med-Cordex database and the size of atmosphere-ocean coupled simulations, not published as of today.
The climate change risk analysis for the islands was undertaken following the concept of impact chains. It aims not only to evaluate the risk, but also to monitor the risk components (vulnerability, exposure) that evolve with time and respond to human interventions. The final objective is to achieve a standardized risk score that allow comparison between islands and decision making on adaptation. From a wider perspective, when European islands are compared to other competing tourism destinations, it is worth reflecting on the overall risk they are expected to face.

Tourists visiting the islands were surveyed in order to analyse their attitudes towards potential climate change impacts and preferences for adaptation policies that can be implemented. Discrete choice experiments were conducted with 2528 tourists visiting 10 European islands. Although tourism managers cannot change the climate, they can nevertheless utilise this knowledge to anticipate which impacts affect the most international tourism arrivals and tourism seasonality, and the economic impact of certain adaptation policies.

At this stage, 2538 EU citizens (frequent travellers) were interviewed. They were asked about their disposition to stay at home and the willingness to pay to visit islands being exposed to several climate change risks. Those impacts that will have higher impacts on the economic value of tourism at the destinations are related with the higher risks of infectious diseases and the higher risks of forest fires, while those with lesser impacts on the economic value of tourism are concerned with the higher risks of heat waves and the expected reduction in beach space because of water intrusion.
Current weather conditions posted on several weather forecast were analysed, as well as daily prices posted on Booking.com by hotels in the island. The link between daily temperature and daily price was estimated, and then projected for the increase in the number of days with excessive temperature projected for the future in two scenarios (RCP2.6 and RCP8.5) and in two time horizons (near future, about 2050; distant future, about 2100). This analysis was undertaken in three islands (Corsica, Sardinia and Sicily).
A specific tool (Google Cloud Vision) is used to scan the content of images posted by tourists on Instagram, while they are on holiday in the islands. Instagram geotagged posts from these islands are scanned according to tourists’ publications (identified by a travel-related hashtag, such as #visit #holiday #travel, etc) during summer 2019 (from June to September). The map of the island shows the word cloud stemming from the analysis of the pictures’ content. This analysis as undertaken for five islands.
The effects of increased sea surface temperatures on aquaculture production (tons) were calculated using a lethal temperature threshold by specie, and considering a production share in each island, under two RCP scenarios and time horizons, which correspond to four water temperature increases. Overall, the great heterogeneity of results among islands, scenarios and time horizons highlights the importance of downscaling climate projections and of adapting them to the different economic characteristics of the islands. In this respect, it is also important to underline that economic projections also substantially differ between islands of the same region (Mediterranean Sea; Atlantic Ocean).
The increase in energy demand (GWh/year), needed for cooling  buildings and desalinate seawater is estimated for the islands under different RCP and time horizons. The relationship between the increase of temperature and the increase of energy demand for cooling is underpinned on the indicator denominated cooling degree days (CDD). The indicator used to translate the increasing temperatures into increases of drinking water demand is the so-called Standardised Precipitation-Evapotranspiration Index (SPEI). The SPEI is an extension of the widely used Standardized Precipitation Index (SPEI).
The increased depreciation (amortization) costs caused by Sea Level Rise (SLR) are estimated for the islands’   125 years’ time horizon under different RCP scenarios of climate change. The study allows to measure the new investments that operators may have to undertake to keep ports’ operability in the future. As SLR happens slowly and gradually, estimations cannot be extended to regions where extreme events are likely to increase in the future.
General Equilibrium models have been used to project socio-economic consequences of climate change for the islands. More specifically it was analysed the effects on 14 sectors of economic activity, GDP, consumption, investments and employment. Islands’ Input-output tables have been updated for each island, to have a basis for the detailed analysis of the respective economy. GEM-E3-ISL and GINFORS macro econometric models were employed. The exercise shows the importance of modeling all Islands individually, because the reactions to climate change and the damages in the respective sectors differ widely across islands and from the main countries.
A consultation process with local stakeholders was carried out, aiming to define alternative adaptation pathways that are framed by the geographic and socio-economic characteristics of each island. Five evaluation criteria were utilize: (a) cost efficiency – ability to efficiently address current or future climate hazards/risks in the most economical way; (b) environmental protection – ability to protect the environment, now and in the future; (c) mitigation win-wins and trade-offs – current ability to meet (win-win) or not (trade-off) the island/archipelago’s mitigation objectives; (d) technical applicability – current ability to technically implement the measure in the island; (e) social acceptability – current social acceptability of the measure in the island.