The use of scenarios in climate risk assessments#
Why scenarios?#
What is a scenario?#
A scenario is a strategic planning tool designed to explore “what-if” situations, helping us prepare for problems that might arise in the future. Scenarios can help to understand how external economic, environmental, social and/or political contexts may influence the risk in a region. In the context of the CLIMAAX project, scenarios relate to climate change and socio-economic development. Examples of scenario variables include rainfall and other aspects of the climate, demographic and socio-economic trends and changes, food consumption and prices.
In the context of Climate Risk Assessments (CRA), scenarios support local authorities in understanding how the future may unfold, making them essential for informed decision-making and effective planning. These pages aim to provide guidance for scenario development, further and more detailed information is available in the list of background reading available at the end of this section.
Why explore multiple scenarios?#
A good plan for regional development and management should not only address present risks and impacts, but should also prepare for risks that might arise in the future. Exploration of multiple scenarios can increase resilience as it helps to:
Map different futures. Mapping different future scenarios is important to understand possible climate risks (such as impacts of heatwaves, wildfires, floods, droughts) and the effect of different actions for climate adaptation and mitigation. Exploring multiple scenarios helps to formulate robust and flexible strategies that can adapt to changing climate conditions.
Guide adaptation actions. Insights gained from scenario analysis can guide adaptation actions. This includes implementing low-regret options that offer benefits regardless of how the future unfolds, as well as robust strategies that perform well across various conditions.
Prevent maladaptation. Exploring multiple climate and socio-economic scenarios helps prevent maladaptation by ensuring that actions are effective across a range of future conditions. It avoids under- or over-investing in solutions by identifying strategies that work in various scenarios.
Combining climate and socio-economic scenarios#
The choice of which scenario to prioritize in planning largely depends on the user’s needs and risk tolerance. High-risk scenarios – such as worst-case scenarios – may require more stringent and risk averse policies and preparations, while low-risk scenarios – such as best-case scenarios – offer a more optimistic but possibly less robust approach. Decision makers must balance the amount of risk they are willing to accept with the potential consequences, ensuring that their strategies align with their goals and resources. Scenario development should always aim to be as simple and transparent as possible while still adequately addressing the problems to be solved.
By systematically developing and analyzing multiple climate and socio-economic scenarios, we can better understand potential future risks and develop strategies that are robust across a range of possible futures. This approach enhances our ability to plan effectively, allocate resources wisely, and build resilience against the uncertainties of climate change.
Ideally, the analysis should be carried out for all the scenario combinations, and the selection of which strategy to use should be based on the evaluation of which is best able to cope with all possible future developments. In reality, most analyses are carried out for a limited number of scenarios. The climate risk management strategy that results from this is then analyzed in a ‘scenario analysis’ to test the strategy for its robustness and flexibility in other possible futures.
Climate scenarios in the CLIMAAX Toolbox#
Two types of climate scenarios are currently widely used to explore future climate conditions. These are the Representative Concentration Pathways (RCPs), representing greenhouse gas concentrations, and Shared Socioeconomic Pathways (SSPs), developed around narratives of plausible trends of socioeconomic futures in the 21st century.
Fig. 17 Representation of SSP narratives, RCPs and future socio-economic developments.#
Representative Concentration Pathways – RCPs#
RCPs help us understand a range of potential futures based on different assumptions of future greenhouse gas emissions. This trend is influenced by population growth, economic activity, socio-economic developments, land-use change and binding targets in emission reductions. Combinations of these assumptions lead to different levels of future greenhouse gas emissions which are used in climate models to simulate the climate response.
RCPs are usually described in a set of four pathways that lead to different levels of ‘radiative forcing’ (the extra energy trapped in the Earth system that warms the planet) caused by humans due to the emission of greenhouse gases. The RCPs cover radiative forcing levels of 2.6, 4.5, 6 and 8.5 W/m² compared to the pre-industrial period by the end of the century. They contain relevant information such as greenhouse gas emissions, air pollutants and land use.
RCP 2.6 is the low emissions scenario where we limit human-caused climate change. Carbon emissions reduce to near zero before the end of the century.
RCP 4.5, a medium emissions scenario, represents a future where we include measures to limit human-caused climate change. This scenario requires global carbon emissions to stabilize by end of the century.
RCP 6.0 uses a high greenhouse gas emission rate and is a stabilization scenario where total radiative forcing is stabilized after 2100 by employment of a range of technologies and strategies for reducing greenhouse gas emissions.
RCP 8.5, the high emissions scenario, represents a future where there are few restrictions on emissions. Emissions continue to increase rapidly through this century, and only stabilize by 2250.
The RCPs have been specifically designed as input to climate models and have become fully integrated scenarios, once they are combined with different SSPs (van Vuuren et al. 2011; van Vuuren et al. 2014).
Future socio-economic developments#
Societal factors such as urban development, infrastructure design, land-use decisions as well as policies that increase or decrease future emissions can either amplify or reduce climate-related risks. Poorly planned urban areas will suffer more from extreme weather, while well-designed cities with resilient infrastructure can better withstand these events.
Understanding how a region might develop in the future is crucial for assessing how societal choices may increase or decrease climate risks. Factors such as population growth, economic policies, and building practices can make some areas more vulnerable, while sustainable planning and adaptation measures can reduce risk. By considering these societal drivers alongside climate projections, we can create more effective strategies to minimize future harm.
There are global SSP-related datasets available for several of the above-mentioned factors that can form the basis for the socio-economic scenario analyses. However, at the national level more detailed scenario information may be present or regional scenarios can be defined based on local knowledge.
Fully integrated scenarios – SSPs#
The Shared Socioeconomic Pathways (SSPs) are part of the most recent IPCC6 scenario framework that refer to possible socio-economic developments instead of emissions. They describe plausible major global developments in the form of narratives that together lead to challenges for emission mitigation and adaptation to climate change in the future. Five SSP pathways have been developed by the international scientific community, allowing to explore climate change across a range of very different futures, see Figure 2:
SSP1-1.9 (Sustainability): A low-challenge scenario aiming for carbon neutrality by mid-century, keeping warming near 1.5°C, aligned with the Paris Agreement.
SSP1-2.6 (Middle-of-the-road development): Also sustainable, but with slightly higher emissions, leading to roughly 2°C warming, similar to the older RCP2.6.
SSP2-4.5 (Inequality): Follows historical trends with uneven development, resulting in medium challenges and radiative forcing of 4.5 W/m², similar to RCP4.5.
SSP3-7.0 (Regional rivalry): High challenges with nationalism and conflict, slowing development and leading to medium-high emissions and forcing of 7.0 W/m².
SSP5-8.5 (Fossil-fuelled development): Rapid, fossil-fuel-intensive growth with high challenges, leading to the highest emissions, forcing of 8.5 W/m², and significant warming, similar to RCP8.5.
Fig. 18 Temperature change under the five SSP scenarios used by the IPCC. Image taken from IPCC, 2021.#
Which climate models should be used?#
Multiple climate models#
A global climate model (GCM) is a mathematical representation of the physical and biogeochemical processes that make up the Earth’s climate. GCMs produce simulations of past, current and future large-scale climates.
GCMs show rising global temperatures with amplified warming in the Arctic, enhancement of the global hydrologic cycle (dry places becoming dryer and wet places becoming wetter), and rising sea level. However, climate models can disagree on many results and projections, especially on a local level analysis. For example, for the near-term future, some models might project significant droughts in a specific region, while another projection might show limited change in the same region. However, both are plausible scenarios. Also, models that perform well in the past do not necessarily provide the ‘best’ projections.
One way to analyse the group of projections from GCMs is by averaging the projections per emission scenario. However, averaging also dampens the change signal. Different climate models project different and sometimes opposite changes in temperature and rainfall patterns, leading to a wide range of plausible outcomes. See the example in Figure 3 below showing significantly varying flood projections for individual models (see multiple oblique lines) while average projections lead to ‘seemingly’ limited change in flood volumes. You can explore multiple climatic conditions using this example exercise.
Fig. 19 Changes in 1/100 year return period river discharge for in the Lielupe basin across various climate models. Adapted from Buskop et al. (2024).#
Additional considerations
CMIP vs CORDEX: When exploring future climate scenarios, you can use for example CMIP6 or CORDEX data. CMIP6 provides data from the latest global climate model runs (~50–100 km) while CORDEX is a downscaled version of the older CMIP5 data (~12 km), which tries to better capture small-scale variations. CORDEX can, therefore, be more useful for local and regional studies.
Choice of time horizon: A project’s planning horizon is another important consideration. Over relatively short periods (i.e., the next decade), the range of projected climate change between different SSPs is small. However, after the middle of the century, the climate projections for the scenarios quickly diverge, which could result in widely varying impacts between the scenarios. In the latest IPCC report, 2021–2040 is referred to as near term, 2041–2060 as mid-term, and 2081–2100 as long term, relative to 1850–1900, which is used as a proxy for the pre-industrial period.
Note
Current conditions, including yearly variations, as well as variations in the coming decades do not significantly change until the 2050s. Therefore, we do not recommend to focus on the scenarios for the near term. Scenario choice impacts outcomes from the mid-century onwards only.
Workflow |
Emission Scenarios |
# Climate Models |
Exploration option |
Climatic impact drivers in the example exercise |
|
|---|---|---|---|---|---|
|
River flooding |
RCP4.5, RCP8.5 |
5 models |
Model average or analyse individual models |
|
|
|
Coastal flooding |
RCP8.5 |
5 models |
Model average |
|
|
Heavy rainfall |
RCP2.6, RCP4.5, RCP8.5 |
Complete CORDEX set |
Analyse individual models |
|
|
Heatwaves |
RCP4.5, RCP8.5 |
5 models |
Model average or analyse spread between models |
|
|
Relative droughts |
SSP1-2.6, SSP3-7.0, SSP5-8.5 |
5 models |
Model average |
|
|
|
Agricultural drought |
RCP2.6, RCP4.5, RCP8.5 |
Complete CORDEX set |
Analyse individual models |
|
|
Wildfire (ML) |
RCP4.5, RCP8.5 |
5 models |
Analyse individual models |
|
|
|
Wildfire (FWI) |
RCP2.6, RCP4.5, RCP8.5 |
6 models |
Analyse individual models |
|
|
Snow |
RCP2.6, RCP4.5, RCP8.5 |
Complete CORDEX set |
Analyse individual models |
|
|
Windstorms |
No projections |
Historical events |
Users create storylines by displacing past events |
No future projections used |
A small exercise
Here are some guiding questions that can help you to build socioeconomic scenarios that are regionally relevant for you:
Describe potential societal risk drivers
Envision various ways of how these might change
How do these changes affect current-day risks?
How to develop your own custom scenarios?#
What makes a scenario useful to take along in the CRA? Here we provide a list of characteristics to keep in mind when developing scenarios:
Relevant: Focus on the specific issue or question you are trying to understand. Provide information and develop indicators that supports targeted decision-making and future planning, rather than being too general or off-topic.
Plausible: Ensure the scenario is realistic and credible, based on facts or logical assumptions. It should represent something that could actually happen, even if it’s a projection about the future.
Consistent: Maintain coherence where all parts fit together logically. Facts and assumptions should align without contradictions, ensuring the scenario makes sense as a whole.
Distinctive: Create scenarios that are unique compared to others. This allows exploration of different possibilities or outcomes, supporting your considerations towards various futures rather than repeating the same ideas.
Practical example
Here we provide a few simple examples of how one can build up a scenario matrix. The matrix is a versatile tool for examining how different combinations of these drivers could influence risk levels.
Increased Flooding and Urban Expansion: In a given region, imagine two factors contributing to climate risk: increase in flooding due to intensified precipitation events and urban expansion in the flood plain area. Increased flooding combined with additional urban expansion results in a greater risk than either factor alone, leading to a red colour in the upper right part of the drawing below.
Fig. 20 A two dimensional scenario matrix where combinations of impact drivers can be made to analyse the risk in each scenario (here indicated by the color scheme).#
Dry Spells, population growth and water usage: For this example imagine three factors influencing climate risks: dry spell length, population growth, and water use per person. Increased dry spells coupled with higher water use and population growth in the region can significantly affect water supplies. Even more than if only one factor was present. This is illustrated in Fig. 21 below: the larger arrows indicate a worsening of the climate impact driver.
Fig. 21 A three dimensional scenario matrix where combinations of impact drivers can be made to analyze the risk in each scenario (here indicated by the color scheme).#
The scenario matrix helps understanding not only isolated impacts but also how combinations of risk drivers might interact, compound, or even mitigate each other, thereby supporting robust decision-making and adaptation planning across various plausible futures.
Storylines of past events#
We can learn from historical events and how these might impact society if they occur again in a different form. Even before diving into future climates, we can already learn from events in the past as their small scale processes such as intense rainfall episodes are often not available in current climate models. Event-based storylines put emphasis on the qualitative understanding of the driving factors involved and the plausibility of those factors rather than quantitative precision. ‘Storylines as a physically self-consistent unfolding of past events’ (Shepherd et al., 2018) bring together the experiences of how an extreme event was experienced in a descriptive narrative that can be used to be better prepared for the future. The use of past events as baseline for storyline development assures a credible justification for the selection of events and provides highly relevant context of the already experienced risks. Moreover, storylines can help to assess future changes in compound and cascading risks.
Intensification of Events: Data from historical events can be modified to replicate what would happen in a warmer world. One could intensify rainfall events, intensify droughts or hurricanes or consider increased sea level.
Displacement of Events: One way to learn from events is to examine historical events that occurred in other regions but could happen locally due to changing climate patterns or just by randomness of where the event actually occurred.
Consecutive Events: A useful exercise can be to consider the impact of multiple events occurring in succession, such as back-to-back windstorms and flooding or simultaneously such as multiple wildfires in different locations. These can help to identify bottlenecks.
Exercise#
How do we put the knowledge provided above into practice? In this exercise, we will explore the range of potential climate futures for your region and help you discover why these changes occur. These plausible changes are highlighted using individual climate model projections that can be reviewed for future weather patterns in one of the other CLIMAAX risk workflows. To help guide you we have worked out an example. We introduce each of the steps using a real-life example in Latvia. Be sure to expand the Latvian example section if you want to see how we execute and interpret each step for a real study.
Define locally relevant climatic impact drivers
Collect observation trends
Collect simulation trends
Look into IPCC reports why this occurs
Latvian example
For our example, we take the Lielupe basin, covering parts of Lithuania and Latvia. For Latvia, the basin has been characterised as a flood zone of national importance. See the figure below for an indication of the flood challenges in the region. Information on how these floods will develop in the future is of significant value as more and more adaptation plans and investments will need to be made as we head into the mid-century.
Fig. 22 Flood extent in the Lielupe Basin#
1. Define locally relevant climatic impact drivers#
Begin by identifying the specific climatic conditions that contribute to your climate-related challenge. The more specific you can be, the easier it will be to track changes and predict future hazards. Below is a set of key climate variables that can be explored on a yearly basis or for specific seasons.
Mean Temperature
Minimum Temperature
Minimum of Minimum Temperature
Frost Days
Heating Degree Days
Maximum Temperature
Maximum of Maximum Temperature
Days with Temperature > 35°C
Days with Temperature > 35°C (Bias Corrected)
Days with Temperature > 40°C
Days with Temperature > 40°C (Bias Corrected)
Cooling Degree Days
Total Precipitation
Maximum 1-Day Precipitation
Maximum 5-Day Precipitation
Consecutive Dry Days
Standardized Precipitation Index (6 months)
Total Snowfall
Surface Wind Speed
Latvian example
Flooding in Latvia has two main flood drivers in Latvia, flooding is primarily driven by two factors. One is the snowpack accumulated in winter that rapidly melts and the melt water flows into rivers during spring. Another driver is spring precipitation, which saturates the soil and when this soil is confronted with extra rainfall, the region starts to flood.
To analyse developments for the two flood mechanisms we select the following climate impact drivers:
Average precipitation March till May
Snowfall in December till February
2. Collect observation trends from interactive climate atlas#
To get a better understanding of what climate trend we can expect we will explore what climate trend is already observed. But before we begin we need to define our region using the IPCC climate zones. The IPCC divides Europe into four climatic zones. These zones are based on areas with similar climate typologies. For each of these zones, climate projection statistics are available.
Fig. 23 IPCC climate zones in Europe. IPCC Atlas#
NEU - Northern Europe
WCE - West and Central Europe
MED - Mediterranean
EEU - Eastern Europe
We analyse past climate behaviour by following the next steps:
Select your region
Select the historical observation data set “era5-land”
Select climate impact drivers and season of interest
Click on “Regional Information” for more information
Spot potential trends
Latvian example
As Latvia is situated on the border between NEU and WCE we need to choose which region. Since the European risk assessment considers the Baltic states to be part of NEU we also choose this region. The graphs below are obtained frpm https://atlas.climate.copernicus.eu/atlas. Overall, there has been an upward trend in ERA5 data for precipitation as can be seen from the graph.
Fig. 24 Precipitation trend in spring for NEU#
In the whole of NEU we find that snow has a slightly increasing trend. The increasing rainfall trend might have contributed to more snow accumulation than temperature increases have reduced snow accumulation.
Fig. 25 Snow trend in winter in NEU#
Based on what we know of the climatic impact drivers we would expect that there are slight increases in both snowmelt and saturated soil floods.
3. Collect simulation trends#
Go to the climate scenarios dashboard to collect information on simulation trends.
Latvian example
The figure below is obtained by selecting the change in Total Precipitation for the season Mar-May for the year 2041 for the NEU region. We notice that that the individual model projections span a much larger range of change than the SSP averaged scenarios. Here we see that individual models behave differently towards the same emission scenarios. This can be either due to differences in the way they represent physical processes or because they are subject to randomness of the weather and might have hit a cold snap or dry spell in their simulations.
For precipitation we notice that when emissions increase SSP model mean changes increase. Individual model projections all indicate an increase in spring precipitation, yet the amount is more unclear. When analysing what could happen in the future we see that in each SSP small increases can occur (<5%) while high increases (>15%) only occur under high emission scenarios. When using this information with known flood mechanisms where soils are saturated and extra rain leads to high discharges, we can expect more of these events to occur. An increase of 10% extra precipitation in the spring season is not unlikely across models and can lead to severe changes in discharges.
Snowfall is expected to decrease when emissions, and thus also temperatures increase. Given that Northern Europe historically received roughly 1.3 mm/day of snow in winter the range given is between a 24% reduction to a 7% increase in snowfall. Few models project an increase and this might only be due to randomness of weather. However, 7% extra snowfall in winter snowpack can result into large changes in melt discharges and might be worth investigating as a potential ‘what if’ scenario. A reduction of 24% is also interesting as it greatly limits snow accumulation and therefore the snow melt flood mechanism.
Fig. 26 Plot of uncertainties for climate variable in NEU#
Fig. 27 Plot of uncertainties for climate variable in NEU#
4. Look into IPCC reports why this occurs#
To get a better sense of why there is such a large range of what could happen in the future we can dive into the IPCC reports to give some clarity. In the reports a scientific background is given to many of the projected changes. To get started we can use ChatClimate. This is an AI tool that is specialised in the IPCC report and guides users towards the chapters and information they need. A good starting point to ask questions is to use the following template.
“What are the expected changes in [region] for [climate impact variable] and why do these changes occur? Also include reasons for differences across climate models.”
if you want to know more or do not understand parts of the answers follow up by using.
“What do you mean by this? [copy text from anwer you want to know more about]”
One can also dive into the IPCC reports and search for explanations. Here are some useful links:
IPCC Climate Change 2021: The Physical Science Basis (for a chapter overview)
Chapter 11: Weather and Climate Extreme Events in a Changing Climate
Chapter 12: Climate Change Information for Regional Impact and for Risk Assessment
Latvian example
Here we dive into the details of why precipitation is expected to increase in Northern Europe and why snowfall is decreasing. Using the information below we have a good basis to dive into the reports if needed or reach out to other experts to tell us more.
Precipitation
When asking ChatClimate: “What are the expected changes in NEU for average precipitation amounts in spring and why do these changes occur? Also include reasons for differences across climate models.” it gives the following answer.
“The expected changes in average precipitation amounts in NEU (North-Eastern Europe) for spring are not explicitly mentioned in the provided information. However, it is reported that in the winter half year, NEU is likely to have increased mean precipitation associated with increased atmospheric moisture and moisture convergence, and intensification in extratropical cyclone activity (Page: 72, Reference: IPCC_AR6_WGI_Atlas). The changes in precipitation are driven by the increase in atmospheric water vapour (about 7% per 1°C of warming near the surface), although this is modulated by various dynamical changes. Changes in average precipitation are also driven by slower processes that constrain future changes over the globe to only 2-3% per 1°C of warming near the surface (Page: 96, Reference: IPCC_AR6_WGI_Chapter11). The differences across climate models are due to pronounced internal variability in this region, particularly NAO (North Atlantic Oscillation) and AMO (Atlantic Multidecadal Oscillation). The spread in climate model projections is still substantial, partly due to these factors (Page: 72, Reference: IPCC_AR6_WGI_Atlas). Additionally, near-term projected changes in precipitation are uncertain mainly because of natural internal variability, model uncertainty, and uncertainty in natural and anthropogenic aerosol forcing (Page: 33, Reference: IPCC_AR6_WGI_Chapter04). However, without more specific information about spring precipitation in NEU, there is not enough info to answer the question fully.”
When asking “What do you mean by this? The differences across climate models are due to pronounced internal variability in this region, particularly NAO (North Atlantic Oscillation) and AMO (Atlantic Multidecadal Oscillation). The spread in climate model projections is still substantial, partly due to these factors (Page: 72, Reference: IPCC_AR6_WGI_Atlas).” We learn more.
“The statement refers to the variability in climate model projections due to internal climate factors, specifically the North Atlantic Oscillation (NAO) and the Atlantic Multidecadal Oscillation (AMO). These are modes of climate variability that can significantly influence regional climates. The NAO, for instance, is the main mode of climate variability over the North Atlantic-European region in all seasons, affecting temperature, precipitation, wind patterns, and even the occurrence of extreme weather events (Page: 4, Reference: IPCC_AR6_WGI_AnnexIV). It can also lead to significant uncertainty in future changes of regional climate, as seen in the CMIP5 RCP8.5 scenario where an increased amplitude and frequency of the NAO during winter is associated with higher precipitation in northern Europe and lower precipitation in southern Europe (Page: 81, Reference: IPCC_AR6_WGI_Chapter08). The AMO, while not explicitly described in the provided information, is another mode of climate variability that can influence climate model projections. The spread in climate model projections refers to the range of possible outcomes predicted by different climate models. This spread is substantial, meaning there is a wide range of possible future climates predicted by these models, partly due to the influence of internal variability factors like the NAO and AMO (Page: 72, Reference: IPCC_AR6_WGI_Atlas). However, it’s also noted that some climate models may have deficiencies that cause the predictable signal associated with the NAO to be weaker than in nature, while others, such as that linked to the SAM, are more consistent with observations (Page: 12, Reference: IPCC_AR6_WGI_Chapter04). This could also contribute to the spread in climate model projections (In-house knowledge).”
Snowfall
When asking ChatClimate: “What are the expected changes in NEU for average snowfall amounts in winter and why do these changes occur? Also include reasons for differences across climate models.”
“The expected changes in average snowfall amounts in winter in North-Eastern Europe (NEU) are likely to decrease due to global warming. This is because regional warming is virtually certain to extend the observed downward trends in snow accumulation, snow water equivalent, and length of the snow cover season in NEU (Page:76, Reference:IPCC_AR6_WGI_Atlas). This is supported by regional and global multi-model and/or single-model ensemble projections including CMIP5, PRUDENCE, ENSEMBLES, and EURO-CORDEX. The changes are attributed to changes in the snowfall fraction of precipitation and to increased snowmelt (Page:76, Reference:IPCC_AR6_WGI_Atlas). The differences across climate models can be due to the different ways they represent physical processes and their interactions. For example, Terzago et al. (2017) showed a large positive bias in the amplitude of the annual snow cycle of EURO-CORDEX 0.11° simulations driven by GCM projections, while reanalysis-driven RCMs showed good agreement with in situ observations (Page:76, Reference:IPCC_AR6_WGI_Atlas). Furthermore, changes in precipitation extremes (including both rainfall and snowfall extremes) can differ drastically from those in average precipitation. The intensification of extreme precipitation is driven by the increase in atmospheric water vapour (about 7% per 1°C of warming near the surface), although this is modulated by various dynamical changes. In contrast, changes in average precipitation are driven not only by moisture increases but also by slower processes that constrain future changes over the globe to only 2-3% per 1°C of warming near the surface (Page:96, Reference:IPCC_AR6_WGI_Chapter11). In summary, the expected changes in NEU for average snowfall amounts in winter are likely to decrease due to global warming, and the differences across climate models can be attributed to the different ways they represent physical processes and their interactions (In-house knowledge).”
Background reading#
Buskop, F. E.; Sperna Weiland, F.; van den Hurk, B. (2024). Amplifying exploration of regional climate risks: Clustering future projections on regionally relevant impact drivers instead of emission scenarios. Environmental Research: Climate, 3(4), 045030. DOI: 10.1088/2752-5295/ad9f8f.
Ebi, Kristie L.; Kram, Tom; van Vuuren, Detlef P.; O’Neill, Brian C.; Kriegler, Elmar (2014): A New Toolkit for Developing Scenarios for Climate Change Research and Policy Analysis. In Environment: Science and Policy for Sustainable Development 56 (2), pp. 6–16. DOI: 10.1080/00139157.2014.881692.
Moss, Richard H.; Edmonds, Jae A.; Hibbard, Kathy A.; Manning, Martin R.; Rose, Steven K.; van Vuuren, Detlef P. et al. (2010): The next generation of scenarios for climate change research and assessment. In Nature 463 (7282), pp. 747–756. DOI: 10.1038/nature08823.
O’Neill, Brian C.; Kriegler, Elmar; Ebi, Kristie L.; Kemp-Benedict, Eric; Riahi, Keywan; Rothman, Dale S. et al. (2017): The roads ahead. Narratives for shared socioeconomic pathways describing world futures in the 21st century. In Global Environmental Change 42, pp. 169–180. DOI: 10.1016/j.gloenvcha.2015.01.004.
Riahi, Keywan; van Vuuren, Detlef P.; Kriegler, Elmar; Edmonds, Jae; O’Neill, Brian C.; Fujimori, Shinichiro et al. (2017). The Shared Socioeconomic Pathways and their energy, land use, and greenhouse gas emissions implications: An overview, Global Environmental Change, Volume 42, 2017, Pages 153-168, ISSN 0959-3780, DOI: 10.1016/j.gloenvcha.2016.05.009.
Reimann, L., Vafeidis, A. T., & Honsel, L. E. (2023). Population development as a driver of coastal risk: Current trends and future pathways. Cambridge Prisms: Coastal Futures, 1, e14. DOI: 10.1017/cft.2023.3.
Shepherd, T.G., Boyd, E., Calel, R.A., Chapman, S.C., Dessai, S., Dima-West, I.M., et al., 2018. Storylines: an alternative approach to representing uncertainty in physical aspects of climate change. Clim. Change 151, 555–571. DOI: 10.1007/s10584-018-2317-9.
van Vuuren, Detlef P.; Edmonds, Jae; Kainuma, Mikiko; Riahi, Keywan; Thomson, Allison; Hibbard, Kathy et al. (2011): The representative concentration pathways. An overview. In Climatic Change 109 (1-2), pp. 5–31. DOI: 10.1007/s10584-011-0148-z.
van Vuuren, Detlef P.; Kriegler, Elmar; O’Neill, Brian C.; Ebi, Kristie L.; Riahi, Keywan; Carter, Timothy R. et al. (2014): A new scenario framework for Climate Change Research. Scenario matrix architecture. In Climatic Change 122 (3), pp. 373–386. DOI: 10.1007/s10584-013-0906-1.