Enter Zoom Meeting


Detecting and attributing climate change: trends, extreme events, and impacts

Detecting and attributing anthropogenic climate change in long-term observed climatic trends is an active area of research, seeking to identify ongoing changes in the climate system, and to quantify the contributions of various external forcing to these changes. Attributable trends, as well as a variety of other emerging constraints, can also be used to constrain climate projections. This science is better established for temperature related variables than for other climate indicator including hydrometeorological variables.

Complementary to this, assessing the extent to which extreme weather events, including compound events, are attributable to anthropogenic climate change is a rapidly developing science, with emerging schools of thought on the methodology and framing of such studies. Once again, the attribution of hydrometeorological events, is less straightforward than temperature-related events. The attribution of impacts, both for long-term trends and extreme events is even more challenging.

This session solicits the latest studies from the spectrum of detection and/or attribution approaches. By considering studies over a wide range of temporal and spatial scales we aim to identify common/new methods, current challenges, and avenues for expanding the detection and attribution community. We particularly welcome submissions that compare approaches, address hydrometerological trends, extremes, impacts, and/or assess implications of recent trends in terms of future changes – all of which test the limits of the present science.

Including CL Division Outstanding ECS Award Lecture
Convener: Aglae JezequelECSECS | Co-conveners: Aurélien Ribes, Pardeep Pall, Seung-Ki Min, Nikolaos Christidis
| Mon, 23 May, 15:10–18:30 (CEST)
Room E2

Mon, 23 May, 15:10–16:40

Chairperson: Aglae Jezequel

Introduction of first block

Neven-Stjepan Fuckar et al.

Extreme weather and climate events - such as intense heatwaves, prolonged droughts and extensive wildfires - are aspects of the evolution of the climate system that are becoming more frequent and stronger in many parts of the world. Extremes can have substantial environmental and socio-economic impacts depending on vulnerability of exposed population, as well as present infrastructure and assets. In August 2021 extreme heat affected the broader Mediterranean region: on 11 August, a weather station in Syracuse (the birthplace of Archimedes on Sicily), Italy, reached 48.8 °C, the European near-surface temperature record, while Kairouan, Tunisia, reached a record 50.3 °C on the same day. 47.4 °C in Montoro set the national record for Spain on 14 August, while on the same day Madrid had its hottest day on the record with 42.7 °C. This was caused by an extensive heat dome, a large area of high pressure in the upper atmosphere leading to strong downward motion that compresses and heats up air in addition to the contribution from radiative heating. Furthermore, this heating was accompanied by devastating wildfires in several Mediterranean countries. Our study utilises a set of observations and reanalysis products combined with large ensembles of CMIP5/6 simulations to examine the role of anthropogenic drivers in this extreme event. We also use large ensembles of specifically designed historical/factual and natural/counterfactual simulations of EC-Earth3.3 coupled climate model at the standard resolution (T255L91 ORCA1L75) to assesses to what extent anthropogenic forcing modified the probability and magnitude of this event involving conditional perspective of the atmospheric circulation. The preliminary results points to a substantial role of the global climate change in modifying likelihood of this extreme event.

How to cite: Fuckar, N.-S., Allen, M., and Obersteiner, M.: Dynamics and attribution of exceptional Mediterranean heatwave in August 2021, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-10936, https://doi.org/10.5194/egusphere-egu22-10936, 2022.

Anjali Thomas et al.

Extreme temperature events (ETEs) have evolved alongside the warming climate over most parts of the world. This study provides a statistical quantification of how human influences have increased the likelihood and frequencies of ETEs in New Zealand, depending on the synoptic weather type. We use the simulation under pre-industrial conditions (natural scenarios with no rise in greenhouse gases (GHGs)) and present-day conditions (anthropogenic scenarios) from the weather@home regional climate model. The ensembles of simulations under these two scenarios are used to identify how increases in GHG concentrations have impacted the frequency and intensity of ETEs based on their connection to different large-scale circulation patterns derived using Self Organizing Maps (SOMs). Over New Zealand, an average 2-3 fold rise in frequencies of extremes occurs irrespective of seasons due to elevated GHG concentrations with a mean temperature increase close to 1℃. For some synoptic situations, the frequency and intensity of ETEs are enhanced. In particular, for low-pressure centers to the northeast of New Zealand, the frequency of occurrence of daily temperature extremes has increased by a factor of 7 between anthropogenic and natural simulations for the winter season, though these synoptic patterns rarely occur. For low-pressure centers to the northwest of New Zealand, we observe extreme temperatures frequently in both anthropogenic and natural simulations which we attribute to warm air advection from the tropics. The frequency of occurrence of these synoptic patterns has also increased by a factor of 2 between the natural and anthropogenic simulations. For these synoptic states, the extremes are observed in the North Island and along the east coast of the country with the highest temperature along the Canterbury coast and Northland. However, the change between the natural and anthropogenic simulations is largest on the west coast along the Southern Alps.

How to cite: Thomas, A., McDonald, A., Renwick, J., Tradowsky, J., and Bodeker, G.: Rise in the frequency and intensity of extreme temperature events over New Zealand in connection to synoptic circulation features, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-6841, https://doi.org/10.5194/egusphere-egu22-6841, 2022.

Svenja Szemkus and Petra Friederichs

In the subproject CoDEx of the BMBF climXtreme project , we are investigating different data compression techniques to detect and attribute changes in the frequency and severity of extreme weather events in a changing climate. Especially for local processes on the atmospheric mesoscale, climate change signals are often masked by additional variability, resulting in poor signal-to-noise ratios. Therefore, these only become visible when the data are analyzed in compressed form. Our focus is on unsupervised learning approaches such as principal component analysis developed specifically for extremes. We focus on extreme precipitation over Germany and analyze how different data compression techniques can be used in a detection and attribution (D&A)study. Besides others, we use the approach proposed by Cooley and Thibaud (2019) on the decomposition of the tail pairwise dependence matrix, as an analogue to the covariance matrix for extremal dependence. Furthermore, we use a dualtree wavelet transform to study changes in extreme precipitation at different scales and different orientations. A D&A study will provide deeper insight into the effects of climate change on extreme precipitation events. 

How to cite: Szemkus, S. and Friederichs, P.: Climate change detection and attribution in extreme precipitation using compact representations, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-8045, https://doi.org/10.5194/egusphere-egu22-8045, 2022.

Bor-Ting Jong and Thomas L. Delworth

Extreme precipitation, both the occurrence and intensity, over the Northeast United States has significantly increased since the 1990s, evidenced by observations. The most salient increase has happened in the fall season (September to November). Understanding the attribution and projection of long-term trends in regional extreme precipitation is essential to adaptationion planning such as infrastructure upgrade. However, such work is challenging due to uncertainties caused by internal climate variability and the requirement of medium-to-high model resolution as well as ensemble size. In this work, we leverage the newly-developed GFDL (Geophysical Fluid Dynamics Laboratory) SPEAR (Seamless System for Prediction and EArth System Research) models which generate 25-km high-resolution simulations (ten members; SPEAR-HI) and 50-km large-ensemble simulations (30 members; SPEAR-MED) for both historical simulations from 1921 to 2014 and projections for the Shared Socioeconomic Pathway 5-8.5 (SSP585) from 2014 to 2100. We aim to address two related scientific questions using GFDL-SPEAR: (1) what are the factors that have contributed to the increasing autumn extreme precipitation over the Northeast US since 1990s? How much of the increase could be attributed to anthropogenic forcing? (2) when would the increased extreme precipitation in response to forced climate change emerge from the noise of internal climate variability?

Our preliminary results first suggest that higher atmospheric resolution in climate models is critical to facilitate the simulations of regional extreme precipitation. For example, SPEAR-HI can simulate comparable frequency of extreme precipitation over the Northeast US (rain rate > 50 mm/day), compared to the observation; while SPEAR-MED underestimates the frequency. Second, the recent increasing Northeast US extreme precipitation is unlikely due to the warming North Atlantic sea surface temperature, even though the timing of the abrupt increase in extreme precipitation coincided with the timing when the Atlantic Multidecadal Oscillation shifted from a cold to warm phase in the mid-1990s. Our ongoing work focuses on evaluating the attributions from other factors including internal variability, aerosols, and greenhouse gas. Last, we analyze SPEAR-HI SSP585 projections and extended control simulations starting from the year 1850. We estimate that the anthropogenically forced increase in the Northeast US autumn extreme precipitation would emerge from the noise of internal climate variability around the 2040s. However, ongoing work will employ more systematic methods to estimate the time of emergence.

How to cite: Jong, B.-T. and Delworth, T. L.: Using an ensemble of high-resolution climate model simulations to detect, attribute, and project changes in extreme rainfall over the Northeast U.S., EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-13358, https://doi.org/10.5194/egusphere-egu22-13358, 2022.

Q&A extreme temperature and precipitation events

Jakob Zscheischler and Flavio Lehner

Extreme event attribution answers the question whether and by how much anthropogenic climate change has contributed to the occurrence or magnitude of an extreme weather event. It is also used to link extreme event impacts to climate change. Impacts, however, are often related to multiple compounding climate drivers. Because extreme event attribution typically focuses on univariate assessments, these assessments might only provide a partial answer to the question of anthropogenic influence to a high-impact event. We present a theoretical extension to classical extreme event attribution for certain types of compound events. Based on synthetic data we illustrate how the bivariate fraction of attributable risk (FAR) differs from the univariate FAR depending on the extremeness of the event as well as the trends in and dependence between the contributing variables. Overall, the bivariate FAR is similar in magnitude or smaller than the univariate FAR if the trend in the second variable is comparably weak and the dependence between both variables is moderate or high, a typical situation for temporally co-occurring heatwaves and droughts. If both variables have similarly large trends or the dependence between both variables is weak, bivariate FARs are larger and are likely to provide a more adequate quantification of the anthropogenic influence. Using multiple climate model large ensembles, we apply the framework to two case studies, a recent sequence of hot and dry years in the Western Cape region of South Africa and two spatially co-occurring droughts in crop-producing regions in South Africa and Lesotho.

How to cite: Zscheischler, J. and Lehner, F.: Attributing compound events to anthropogenic climate change, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-7829, https://doi.org/10.5194/egusphere-egu22-7829, 2022.

Aditya Narayan Mishra et al.

Between 22-26 June 2009, Austria witnessed a rampant rainfall spell that spread across populated areas of the country. High-intensity rainfall caused 3000+ landslides in southeast Styria, and property damages worth €10 Million in Styria itself. Elsewhere in Austria, flooding amounted to reparations worth €40 Million. Numerous synoptic-scale studies indicated the presence of a cut-off low over central Europe and excessive moisture convergence behind the extreme event. In a warmer climate change scenario, such an extreme precipitation event may manifest into a more intense event due to the higher water holding capacity of air with increased temperatures, but this reasoning may not be so straightforward considering the complex physics of precipitation, more so in a topographically heterogeneous region such as the GAR (Greater Alpine Region).

The flooding and landslides caused in the region raise an alarm and thus motivate this study whereby we investigate if the rainfall event did become stronger with time due to climate change compared to how it would have been in a counterfactual (climate change free) past. Here we have deployed the CCLM high-resolution regional model coupled with a statistical landslide model to simulate this event (rainfall and landslides) in a pseudo (surrogate) warming scenario. A marked decrease in rainfall intensity is observed in the simulations for 1° cooler climate (pre-industrial past) and the consequent landslide risk is reduced varying across GCMs that were used to derive the boundary conditions from.

We discuss the results from the lens of attribution perspective - how conditional attribution is much more useful compared to the conventional risk-based approach of attributing extreme events. The novelty of our approach lies in using a high-resolution convection-permitting regional model for a landslide attribution study.

How to cite: Mishra, A. N., Maraun, D., Truhetz, H., Knevels, R., Bevacqua, E., Proske, H., Petschko, H., Leopold, P., and Brenning, A.: Attribution of 2009 extreme rainfall & landslide event in Austria , EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-5845, https://doi.org/10.5194/egusphere-egu22-5845, 2022.

Manish Kumar Dhasmana et al.

The role of global warming and climate change in altering the likelihood of extreme weather events is becoming increasingly evident. Event attribution refers to the collection of frameworks that use observed data and climate model simulations for quantifying the contribution of human-induced (anthropogenic) climate change in changing the event probability. In this study, we present a multi-method event attribution analysis of the catastrophic extreme precipitation and flooding event in Kerala, India in August 2018, that resulted in widespread destruction and loss of lives. Two methods- (i) based on factual (Historical) and counterfactual (HistoricalNat) runs from 5 CMIP6 climate models, and (ii) based on observed data, scaled to 2018 (factual) and 1901(pseudo-counterfactual) climates, are considered for quantifying the fraction of attributable risk (FAR) of the 2018 event. Using an objective approach, we first define the 2018 event as the 4-day cumulative rainfall over the Periyar river basin (PRB), during 15- 18 August, 2018. This event has a return period of 373 years (90% CI: 72-1200 years). The 1-day maximum streamflow at one of the outlets of the PRB, where maximum impact during the event was reported, is used for attributing the associated flood event. Simulated using VIC hydrological model, the streamflow event is found to have a return period of 34 years (90% CI: 12-286 years). The FAR from the climate model ensembles is -0.18 and -0.14, for the precipitation and streamflow events, respectively. The scaled observations also give negative FARs: -0.97 for precipitation and -0.93 for streamflow. These values imply that the 2018 event is exceptionally less likely due to climate change. In other words, our results underline the definitive absence of anthropogenic role in the 2018 event.

How to cite: Dhasmana, M. K., Mondal, A., and Zachariah, M.: Multi-method attribution of the extreme precipitation and flood of 2018 in Kerala, India., EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-1270, https://doi.org/10.5194/egusphere-egu22-1270, 2022.

Sabine Undorf et al.

Statistically rigorous methods to attribute large-scale, long-term changes in observed climate to anthropogenic forcing are well established and the attribution of individual extreme events has advanced rapidly too. Further attributing climate impacts in society and ecosystems can be important for understanding and assessing loss and damage, for informing adaptation policies, and for motivating both adaptation and mitigation efforts. For this purpose, a counterfactual climate dataset was recently published within the Inter-Sectoral Impact Model Intercomparison Project (Mengel et al., 2021). Constructed by removing the long-term shifts in daily reanalysis data that are correlated to global-mean temperature change, the dataset does not address anthropogenic climate change, but its large spatial and temporal coverage and the range of variables covered as well as minimum requirements on computational tools and data make it a desirable resource for this interdisciplinary problem.

Here, we trial the use of that counterfactual dataset for the quantification of climate impacts on agricultural crop yields, which are of paramount importance to many of the regions most exposed and vulnerable to climate change, not least for food security. We present results from case studies that examine the impacts of selected drought events and are chosen based on reports of substantial food security impacts, on the availability of crop yield data, and on the existence of published scientific literature of a corresponding climate attribution study. The latter allows the comparison with results using methods that isolate the anthropogenic (combined, or by individual forcing) climate change signal. Together with more systematic discussion, this gives an idea of the degree to which our results on the contribution of any climate variations to the observed impacts may be a proxy for the anthropogenic climate change contribution specifically.

Impacts are explicitly simulated using statistical crop models that are established in the agricultural and agronomic literature, built and validated based on the observed record. The use of the single-realisation, weather-preserving factual and counterfactual dataset gives a deterministic rather than probabilistic estimate, but parametric and structural crop-model uncertainty is characterised, and the robustness of the results to different observational crop-yield datasets assessed. We collaborate with local stakeholders to ensure appropriate consideration of non-climatic factors and to improve data availability and quality. Our work combines perspectives of climate attribution, disaster risk reduction, and agricultural science to enhance attributing loss and damage in agriculture to climate change.

How to cite: Undorf, S., Schauberger, B., and Gornott, C.: Attributing observed climate change impacts in agriculture using observationally-derived counterfactual climate data and statistical crop-yield modelling, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-4326, https://doi.org/10.5194/egusphere-egu22-4326, 2022.

Zhongwei Liu et al.

In response to the occurrence of a number of large wildfire events across the world in recent years, the question of the extent to which climate change may be altering the meteorological conditions conducive to wildfires has become a hot topic of debate. Despite the development of attribution methodologies for extreme events in the last decade, attribution studies dedicated explicitly to wildfire, or otherwise extreme ‘fire weather’, are still relatively few. In turn, there is a lack of consensus on how to define fire risk in a meteorological context, posing a challenge for research in this subfield. Recent work has offered clarification on uncertainties associated with the choice of meteorological indicator to represent fire weather in the context of extreme event attribution but there are additional sensitivities that are still not fully understood.

Here, using established statistical methodologies applied to six large (>10-member) CMIP6 model ensembles, we conduct probabilistic attribution of fire weather extremes across the world’s fire-prone regions. We assess trends in extremes in the Canadian Fire Weather Index (FWI) using extreme value distributions, fitted with both annual maxima and peaks over a pre-defined threshold, and scaled to global mean surface temperature. An initial evaluation of model performance shows that, while all models are able to reasonably reproduce observed global patterns in extreme distribution parameters, there are some notable differences at the regional scale. Subsequently, we use probability ratio maps to quantify the influence of rising global temperatures on the changing frequency of FWI extremes. Our results highlight the sensitivity of probabilistic fire weather attribution methodologies to the choice of climate model ensemble. In conclusion, we therefore make a set of recommendations for future attribution of extreme fire weather episodes: (i) the use (and comparison) of multiple model ensembles; (ii) robust evaluation of model capacity to represent fire weather extremes.

How to cite: Liu, Z., Eden, J., Dieppois, B., Drobyshev, I., and Blackett, M.: Identifying sensitivities and uncertainties in the attribution of global fire weather extremes using CMIP6 ensembles, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-10653, https://doi.org/10.5194/egusphere-egu22-10653, 2022.

Q&A Attribution of compound events

Fraser Lott et al.

Probabilistic event attribution aims to quantify the role of anthropogenic climate change in altering the intensity or probability of extreme climate and weather events. It was originally conceived to calculate the costs associated with any increased likelihood of the meteorological event in question. However, only recently have such studies attempted to divide liability between polluting nations and ascribe a cost. Recent protests indicate a perception that older generations have the greater responsibility for climate change. In this paper, we examine how a portion of the cost of an event can be attributed to any individual person, according to their age and nationality. We demonstrate that this is quantitatively feasible using the example of the 2018 summer heatwave in eastern China and its impact on aquaculture. A relatively simple technique finds sample individuals responsible for between 0.53 and 18.10 yuan, increasing with their age and their country’s emissions over their adult lifetime since the first international consensus on carbon emissions was reached. This provides an illustration of the scale of such responsibilities, and how it is affected by national development and demographics. Such data can support decisions, at national and international levels, on how to fund recovery from climate impacts. It offers a simple quantitative approach for individuals to know their impact on the climate, or for governments to use in making policy decisions about how best to distribute costs of climate change.

How to cite: Lott, F., Ciavarella, A., Kennedy, J., King, A., Stott, P., Tett, S., and Wang, D.: Quantifying the contribution of an individual to making extreme weather events more likely, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-705, https://doi.org/10.5194/egusphere-egu22-705, 2022.

Eduardo Alastrué de Asenjo et al.

The Atlantic Meridional Overturning Circulation (AMOC) is a crucial feature of Earth’s climate, and the possible recent AMOC weakening in association with anthropogenic emissions is a topic of current scientific discussion. To assess whether anthropogenic emissions affect the circulation, we examine low AMOC strengths with the framework of the causal counterfactual theory.  

We compare the occurrence of low strengths in a factual world with all forcings present against a counterfactual world, equivalent except for anthropogenic forcings. To represent these two worlds, we use past and future 30-member ensembles of simulations from the Max Planck Institute Earth System Model (MPI-ESM1.2-LR), whose AMOC characteristics at 26.5°N are in overall agreement with observations from the RAPID program.

Considering the whole historical period (1850-2020), our results show that causation probabilities for anthropogenic forcings on low AMOC strengths at 26.5 °N are generally low. However, within future projections, we find that anthropogenic forcings will very likely be a necessary cause of any AMOC strength below 16.0 Sv by about 2040. These results are sensitive to the choice of attribution window: if the starting year is chosen closer to the present, all probabilities substantially increase. Within the analyzed simulations, the main attribution results found at 26.5°N are confirmed for other latitudes in the North Atlantic.

How to cite: Alastrué de Asenjo, E., Barkhordarian, A., Brune, S., and Baehr, J.: Causal attribution of low AMOC strengths to anthropogenic influence, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-2162, https://doi.org/10.5194/egusphere-egu22-2162, 2022.

Elizaveta Malinina et al.

In British Columbia, and in the Pacific North West America in general, 2021 was a year fraught with extreme weather events. First, in late June 2021 an unprecedented heat wave claimed over 600 lives, followed by in an extreme flood in mid-November that became one of Canada’s one of most expensive natural disasters. In this study, we provide separate attribution of these two events using CMIP6 temperature and precipitation data. For both events, we compare the current climate data with that from prior decades as well as data for the end of the 21st century under the SSP2.45 scenario. By fitting generalised extreme value distributions to all datasets, we were able to quantify the risk ratios as well as provide analysis of the changing frequency of such extreme events with in a warming climate.

How to cite: Malinina, E., Gillett, N., Kirchmeier-Young, M., Zhang, X., Anslow, F., and Zwiers, F.: Attribution of 2021 Weather Extreme Events in British Columbia, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-10422, https://doi.org/10.5194/egusphere-egu22-10422, 2022.

Péter Szabó et al.

Although anthropogenic global warming is well-known within the scientific community, the public is still not certain how to associate specific local weather and climate events to this issue. Therefore, it is essential to raise public awareness by providing sound and readily understood scientific information, and to explain how humans contribute to specific (extreme) events. Although a few case studies for past high-impact weather events were assessed in Hungary, a systematic analysis of long-term past and future trends of these indices is missing, and their linkage to anthropogenic activity has not been addressed at all. Our attribution project (started in September 2021) aims to fill this gap in Hungary: the results of the analysis of seasonally relevant indices are published in each season at the time of an (extreme) event occurrence. The dissemination is done via an already established Hungarian platform (https://masfelfok.hu/) reaching the public with readily understood climate change information through their broad media coverage and a large social media network.

The assessments are prepared within the project using several data sources: (1) an ensemble of CMIP6 global climate model simulations of both natural-only forcings and historical runs, (2) an ensemble of regional climate model simulations from Euro-CORDEX, including both RCP4.5 and RCP8.5 scenarios, (3) a fine-resolution, homogenized observation-based gridded data for Hungary, (4) the ERA5 reanalysis. We address seasonally relevant extreme and compound events, and the attribution of pre-selected indices to anthropogenic activity through their intensity, duration, and frequency changes. For instance, frost days, annual temperature minima and return values, snowfall and heavy snowfall days are evaluated for winter, while the start of vegetation period, late frost, dry and heavy precipitation days for spring, as they are of most public interest. Furthermore, we determine how many seasonal record low/record high breakings happen and how large area within the domain is affected.

Preliminary winter results suggest that the decrease of frost days in Hungary is clearly due to anthropogenic activity, while it is not the case for the annual minimum temperatures but will be in the future. No significant decrease has been detected for the snowfall and heavy snowfall days, and the effects of anthropogenic activity on these indices will only occur following the pessimistic future scenario.

How to cite: Szabó, P., Bartholy, J., Barna, Z., Bokros, K., Bordi, S., Mráz, A., Pieczka, I., and Pongrácz, R.: Attribution of seasonally relevant winter and spring climate indices in Hungary, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-9707, https://doi.org/10.5194/egusphere-egu22-9707, 2022.

Q&A & closing of first block

Mon, 23 May, 17:00–18:30

Chairperson: Nikolaos Christidis

Introduction of second block+ laudation

CL Division Outstanding ECS Award Lecture
Marlene Kretschmer

Due to their relevance for regional weather and climate, teleconnections are an extremely active area of research. One key task is to quantify the contribution of a teleconnection to regional anomalies in both models and observations. This is, for instance, important to improve forecasts on time scales ranging from subseasonal to multidecadal, or to attribute ensemble spreads to changes in large-scale drivers. However, robustly estimating the effects of a teleconnection remains challenging due to the often simultaneous influences of multiple climate modes. While physical knowledge about the involved mechanisms is often available, how to extract a particular causal pathway from data are usually unclear.

In this talk I argue for adopting a causal inference-based framework in the statistical analysis of teleconnections to overcome this challenge. A causal approach requires explicitly including expert knowledge in the statistical analysis, which allows one to draw quantitative conclusions. I illustrate some of the key concepts of this theory with simple examples of well-known atmospheric teleconnections. Moreover, I show how the deductive nature of a causal approach can help to assess the plausible influence of Arctic sea ice loss on mid-latitude winter weather, thereby helping to reconcile differences between models and observations. I finally discuss the particular challenges and advantages a causal inference-based approach implies for climate science.



Kretschmer, M., Adams, S. V., Arribas, A., Prudden, R., Robinson, N., Saggioro, E., & Shepherd, T. G. (2021). Quantifying Causal Pathways of Teleconnections, Bulletin of the American Meteorological Society, 102(12), E2247-E2263. Retrieved Jan 13, 2022, from https://journals.ametsoc.org/view/journals/bams/102/12/BAMS-D-20-0117.1.xml

Kretschmer, M., Zappa, G., and Shepherd, T. G. (2020), The role of Barents–Kara sea ice loss in projected polar vortex changes, Weather and Climate Dynamics, doi: 10.5194/wcd-1-715-2020

How to cite: Kretschmer, M.: Quantifying Causal Pathways of Teleconnections, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-8246, https://doi.org/10.5194/egusphere-egu22-8246, 2022.

Q&A - medal lecture

Andrew Schurer et al.

The latest generation of climate models (CMIP6) have very different large-scale surface temperature variability and inconsistencies with observed climate have been found in the variability in several regions. Given that detection and attribution, in common with many climate analyses, relies on model internal variability for uncertainty ranges, it is crucial to better constrain this variability. Here, we compare the latest climate models to observed variability to determine where and on what timescales discrepancies occur, with the models found to be, in general, too variable on annual timescales over land and with not as much variability as the observations particularly over the Southern oceans at multi-decadal timescale. We further use paleo-proxy reconstructions, supported by observational datasets finding that the majority of models have variability consistent with large-scale mean temperature on multi-annual and multi-decadal timescales. Finally, the presentation will explore the implications of these findings on key detection and attribution analyses, in particular the attribution of warming since pre-industrial times to anthropogenic forcings.

How to cite: Schurer, A., Luecke, L., and Hegerl, G.: Constraining internal surface temperature variability and its implications for detection and attribution, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-5193, https://doi.org/10.5194/egusphere-egu22-5193, 2022.

Olivia Ferguglia et al.

Climate models are fundamental tools to understand the complexity of the climate system, to study  the processes at work and to provide credible future climate projections. Unfortunately, models often disagree significantly in the amplitude of different climate change signals and in their representation of the role of important feedbacks. In the past years the “Emergent Constraints” methodology has been developed for reducing uncertainties in climate-change projections. An Emergent Constraint (EC) is a statistical relationship between the inter-model spread of a measurable aspect of the present-day climate (predictor) and the inter-model spread of a variable projection (predictand), under a climate change scenario. If a significant correlation is found, observations of the predictor can be used to constrain model projections of the predictand and the uncertainties in climate model outputs can be narrowed. 

In the last two decades, ECs have been identified in different branches of climate science although just a limited number of these ECs is related to the hydrological cycle. Recently, a relevant number of EC in the literature was discovered to lack a satisfying physical explanation and many, developed and tested with CMIP5 ensemble, seem to be not-significant in the new CMIP6 ensemble. However, the analysis regarding ECs related to the hydrological cycle is still incomplete. The aim of this work is to test three ECs related to mean-precipitation and extreme precipitation events, originally identified in CMIP3 or CMIP5 data, and to evaluate if their statistical significance survives also in the CMIP6 ensemble: (a) global hydrological sensitivity used to constrain future changes in local extreme precipitation: we find this relationship not to be robust in CMIP6 models; (b) future changes in the Indian summer monsoon precipitation, constrained by Western Pacific mean precipitation: this relationship is not robust with the new ensemble; (c) changes in future extreme tropical precipitation, constrained by the same variable calculated in the past: we find this EC to be robust both in CMIP5 and CMIP6.

How to cite: Ferguglia, O., Palazzi, E., and von Hardenberg, J.: Robustness of precipitation Emergent Constraints in CMIP6, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-9497, https://doi.org/10.5194/egusphere-egu22-9497, 2022.

Andrew Ballinger et al.

The North Atlantic Oscillation (NAO) is the leading mode of climate variability over the North Atlantic region, affecting temperature and rainfall over timescales from days through to seasons and decades. Previous studies have shown that variations in the NAO yield significant multi-decadal trends in European rainfall, especially in winter, and the magnitude of past multi-decadal NAO variability is generally not reproduced by CMIP class models This has important implications for deriving observational constraints, and the application of these scaling factors to projections of future European rainfall.

We have constructed two sets of multi-model-mean spatiotemporal fingerprints of European rainfall: one set that retains NAO variability, and another set that excludes the variability associated with the NAO (removing it using a simple regression technique). Following the so-called Allen-Scott-Kettleborough ‘ASK’ method, we conduct total-least-squares regressions using the two different sets of fingerprints against the observations in order to analyse the impact of removing the NAO in potentially enhancing the signal-to-noise. The derived scaling factors shed light on the ability of CMIP6 models to reproduce the magnitude of the forced response in precipitation, and confidence intervals for each of the scaling factors describes the range of magnitudes of the model response that are consistent with the observed signal.

Here we focus on one clear example, northern European rainfall, although we have also analysed additional European regions and surface air temperature across all of the seasons. There is an increasing trend in observed rainfall anomalies associated with the NAO over northern Europe, most pronounced in winter, which is not replicated in models. Once the variability associated with the NAO is removed the magnitude of the observed trends is reduced, and the scaling factors (derived in the ASK framework) are similarly reduced.

Along with a shift in the magnitude that comes from the modified observations, the constraint also tightens (the spread in consistent scaling factors narrows) due to the increased signal-to-noise in the modelled response once the NAO is removed from the simulations. This has important implications for the use of scaling factors to constrain future projections of European climate. The observed decadal to multi-decadal trends resulting from known modes of internal variability should be accounted for in the derivation of scaling factors to better capture the forced signal and bolster confidence in the constrained projections.

How to cite: Ballinger, A., Schurer, A., and Hegerl, G.: Accounting for the NAO when applying observational constraints to future European climate projections, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-12074, https://doi.org/10.5194/egusphere-egu22-12074, 2022.

Romana Beranova and Radan Huth

Since long-term changes in precipitation variability are an important aspect of climate change, it is necessary to know whether and how they are changing. We consider two measures of short-term precipitation variability: (i) dry-to-wet and wet-to-wet transition probabilities; they are sufficient for the description of the other two probabilities (dry-to-dry and wet-to-dry) provided precipitation occurrence follows the two-state first-order Markovian process and (ii) mean duration of dry and wet spells, that is, sequences of days without and with measurable precipitation.

The daily precipitation data are taken from European Climate Assessment and Dataset project. We examine 395 precipitation station series from 1961 to 2010. Long-term trends of seasonal values of variability measures and their statistical significance are calculated by non-parametric methods (Mann-Kendall test, Kendall statistic). We found out that statistically significant trends of transition probabilities are more frequent in winter than in summer. In winter, there are positive trends of wet-to-wet probabilities and negative trends of dry-to-dry probabilities in northern Europe.

How to cite: Beranova, R. and Huth, R.: Trends in short-term precipitation variability, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-4622, https://doi.org/10.5194/egusphere-egu22-4622, 2022.

Julio Isaac Montenegro Gambini and Magaly Cusipuma Ayuque

Monitoring trends and continuous changes to complement local studies on discharge and precipitation are becoming increasingly important, assessing systematically its temporal and spatial variation characteristics. It is also necessary to provide detailed study on the driving mechanisms of those variations, which can be a result of anthropogenic activities as it specifically affects particular regions and terrains. These data play very significant roles in measuring and forecasting potential impacts and lead to future improved regional flow regulations for better water resources management. One of the most effective methods for observing the effects of climate change on hydrometeorological variables is the trend analysis, with recently new graphical methods that represents an innovative alternative to the classical ones. In this work, mean monthly, annual mean, minimum and maximum precipitation was examined to analyze spatiotemporal variations, seasonality shifts and trends with records that can extend from 1910 to 2019 in the hydrometeorological network. The flow discharge were also analysed considering catchments with unimpaired streamflow in unregulated rivers. Different homogeneity and shift detection methods were used to check their homogeneity before conducting trend analysis. Innovative Polygon Trend Analysis (IPTA), Innovative trend analysis (ITA) with the Significance Test and Mann-Kendall (MK) non-parametric methods were compared, showing their sensitivity and their ability to explain a monthly trend sequence and periodicity in the studied region in order to explain adequately the temporal internal variability.

How to cite: Montenegro Gambini, J. I. and Cusipuma Ayuque, M.: Graphical analysis applications to study variability and trends of rainfall and streamflow in Colombian catchments, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-13379, https://doi.org/10.5194/egusphere-egu22-13379, 2022.

Q&A Variability and observational constraints

Iris de Vries et al.

Detection and attribution (D&A) of anthropogenically forced changes to precipitation is challenging due to the high internal variability of precipitation and the limited spatial and temporal coverage of the observational records. These factors result in a low signal-to-noise ratio of potential regional and even global trends.

Here, we use a statistical method – regularised linear regression, or ridge regression – to create physically interpretable detection metrics (fingerprints) for D&A of changes in the precipitation distribution with a high signal-to-noise ratio. The regression coefficients that make up the fingerprints of forced change are based on the CMIP6 multi-model archive data masked to match observational coverage, and are then applied to gridded precipitation observations to assess the degree of forced change detectable in the real-world climate. 

We show that the signature of forced change is detected and attributed to external forcing in two different observational datasets in global metrics of mean and extreme precipitation (PRCPTOT, and Rx1d, respectively). If the global mean trend is removed from the data, forced changes are still detected, indicating that climate change affects the spatial patterns of precipitation, and increasing confidence in the results of this method for D&A of precipitation, as well as in climate models capturing the relevant processes that contribute to the regional patterns of change. Furthermore, we show the sensitivity of our D&A results to several ‘design choices’, including target metric of forced change, regularisation parameter, season of interest, (spatial coverage of) observational dataset used, the forced trend length and the region of interest (tropics vs. subtropics). 

The method is largely insensitive to target metric and regularisation parameter, increasing confidence in the robustness of the results. However, we find that June-July-August generally has low forced trend signal-to-noise ratio in both mean and extreme precipitation. Furthermore, the observational dataset choice affects detectability not only through coverage differences but also dataset disagreement, and the chosen trend length can result in different forced trends when comparing observations to model projections. These sensitivities may explain apparent contradictions in recent studies on whether models under- or overestimate the observed increase in extreme precipitation. Lastly, the detection models are found to rely primarily on the signal in extratropical northern hemisphere data, which is at least partly due to observational coverage, but potentially also due to presence of a more robust signal in the northern hemisphere in general. When these regions are excluded, detection of significant forced changes is no longer possible, which may also have implications for the ability to assess risks and inform adaptation policies in the tropics and the global south.

Ridge regression is powerful for D&A of the precipitation distribution, opening up possibilities for extension of the method to learn more about mechanisms driving forced changes in precipitation. However, internal variability, limited coverage in time and space, and dataset disagreement in precipitation data continue to play a large role.

How to cite: de Vries, I., Sippel, S., and Knutti, R.: Detection of forced changes in the precipitation distribution using ridge regression, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-11239, https://doi.org/10.5194/egusphere-egu22-11239, 2022.

Carla Roesch et al.

Anthropogenic aerosols (AER) have been found to impact both Earth’s energy and water cycle. Like greenhouse gases (GHG) they are an anthropogenic climate forcing, which will play an important role in shaping Earth’s future climate. To improve future predictions, it is, therefore, fundamental to understand and quantify the individual impacts these two forcings have on the climate system. This can be achieved by using detection and attribution methods facilitating the differentiation of the response of the climate system to different forcings.

Separating the signal of individual anthropogenic effects related to greenhouse gas and aerosol emissions is hindered by large uncertainties in the response to aerosol forcing in different climate models. Thus, in this study we investigate a joint change in temperature and precipitation to reduce the signal-to-noise ratio and better constrain the impact of anthropogenic aerosols since 1979. Building on previous findings on how aerosols affect climate, we focus on shifts in tropical precipitation by tracking wet/dry regions as well as changes in the interhemispheric temperature asymmetry (ITA) and the diurnal temperature range (DTR), due to its unique response to different radiative forcings. Individual fingerprints are derived from large-ensembles of historical single-forcing simulations from three models that are part of phase 6 of the Coupled Model Intercomparison Project (CMIP6).

We find inter-model agreement in the trends for wet regions, ITA, and DTR in single-forcing and historical (all-forcing) runs. Contrasting trends in these time series derived for AER-only and GHG-only simulations suggest that aerosols have offset some of the greenhouse gas induced precipitation and temperature changes in the past.  While a drying in the dry regions can be observed for GHG-only simulations, inter-model agreement is not found for aerosols. Early results show an improved constraint on the detection of a greenhouse gas signal when investigating a joint change in wet and dry regions, which is refined by including the other variables and indicators.

How to cite: Roesch, C., Ballinger, A., Schurer, A., and Hegerl, G.: Using temperature and precipitation combined to detect and attribute aerosol effects on large-scale climate, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-8088, https://doi.org/10.5194/egusphere-egu22-8088, 2022.

Gopika Suresh et al.

Estimating when anthropogenically-forced signals emerge from ambient natural climate variability is crucial for climate-change detection. Here we propose a new method for estimating the emergence time of climate signals based on a significance test on the nonlinear trend rather than the commonly-used method based on a signal-to-noise ratio. This method is less sensitive to the choice of a confidence level (0.1, 0.05 or 0.01) than the previous method to commonly-used signal to noise ratios (0.5, 1 or 2). For signals that tend to emerge in the late 21st century, our method tends to yield earlier detection dates, by taking the large number of degrees of freedom into account. Here, we apply this method to relative SST (RSST, SST minus its tropical mean) changes, which tend to emerge much later than SST change signals. RSST is an indicator of changes in the atmosphere vertical stability and thus of changes in tropical cyclones intensity and precipitation. By 2100, CMIP5/6 projections indicate greater than tropical average warming (positive RSST signal) in the central and eastern equatorial Pacific, equatorial Atlantic, and Arabian Sea, and reduced warming (negative RSST signal) in the three southern hemisphere subtropical gyres. In general agreement with observations, relative warming in the Arabian Sea and relative cooling in the South-Eastern Pacific are already detectable in a majority of models (median emergence time < 2020), making these regions suitable for testing a model's ability to predict a regional SST trend. In contrast, RSST signals in other regions do not become detectable until after 2050. Tropical precipitation projections indicate more (less) rainfall in regions of positive (negative) RSST change that typically emerge one or two decades later than RSST signals. This lack of currently-detectable regional rainfall trends in CMIP models makes it difficult to evaluate their ability to predict tropical regional rainfall trends. In general, signals tend to emerge later in CMIP6 than in CMIP5 due to both weaker signals and larger climate noise. The only exception is Sahel, where CMIP6 models already display a detectable rainfall increase, that is not yet detectable in CMIP5.

How to cite: Suresh, G., Suresh, I., Lengaigne, M., Vialard, J., Izumo, T., and Kwatra, S.: When do regional tropical climate signals become detectable in CMIP5/6 simulations?, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-8622, https://doi.org/10.5194/egusphere-egu22-8622, 2022.

Luke Grant

In addition to their changes to carbon pools, land use and land cover changes (LULCC) alter climates biogeophysically through their effects on surface fluxes for energy and water. These resulting climatic changes in temperature manifest differently across study type (observational or model-based) and spatial-temporal scales depending on the predominance of albedo, evaporative fraction and surface roughness as causal factors. With growing future demand for land-exhaustive activities to address societal needs and interest in mitigation strategies involving reforestation/afforestation, it is important to understand how past LULCC contributed to climate change. Here we assess the prevalence of the historical LULCC signal in the warmest average month of daily maximum temperatures using regularized optimal fingerprinting for detection and attribution. We use the simulations of four global climate models from CMIP6 historical and hist-noLu experiments and separate observations from the Climatic Research Unit and Berkley Earth. Aggregating data according to the new IPCC AR6 reference regions and regressing observations onto hist-noLu and lu (historical - hist-noLu) in a 2-way regression, we find that LULCC is not sufficiently detectable at continental and global scales for four GCMs and their multi-model mean. This is confirmed by the nearly unchanging detectability of historical climate change in separate 1-way regressions for historical and hist-noLu. To further explore lu, we confirm the predominance of noise and the non-local effects of LULCC over its local effects by finding insignificant signal-to-noise ratios for the fingerprint of forest cover on lu using a principal component analysis.

How to cite: Grant, L.: Biogeophysical effects of land use and land cover change not detectable in CMIP6 warmest month., EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-2345, https://doi.org/10.5194/egusphere-egu22-2345, 2022.

Lin Xue et al.

Vegetation in Northeast China (NEC) has faced dual challenges posed by climate change and human activities. However, the factors dominating vegetation development and their contribution remain unclear and cannot be precisely discussed. In this study, we conducted a comprehensive evaluation of the response of vegetation in different land cover types, climate regions, and time scales to water availability from 1990 to 2018 based on the relationship between normalized difference vegetation index (NDVI) and Standardized Precipitation-Evapotranspiration Index (SPEI). The effects of human activities and climate change on vegetation development were quantitatively evaluated using the residual analysis method. We showed that the area percentage with a positive correlation between NDVI and SPEI increases with the extension of time scales. NDVI of grass, sparse vegetation, rain-fed crop, and built-up land as well as sub-humid and semi-arid areas (drylands) correlated positively with SPEI, and the correlations increased with the extension of time scales. The negatively correlated area was concentrated in humid areas or areas covered by forests and shrubs. The maximum water surplus period for irrigated crops and forests, shrubs, wetlands, humid areas were 1-month and 6-months, respectively. Vegetation water surplus in humid areas weakens with warming, and vegetation water constraints in drylands enhance. Moreover, potential evapotranspiration had an overall negative effect on vegetation and precipitation is a controlling factor for vegetation development in semi-arid areas. Within the period of study, 53% of the vegetated area in NEC showed a trend of improvement, which is mainly attributed to human activities (93%), especially through the implementation of ecological restoration projects in NEC. The relative role of human activities and climate change in vegetation degradation areas were 56% and 44%, respectively. Our findings highlight that the government should more explicitly consider the spatiotemporal heterogeneity of the influence of human activities and water availability on vegetation under changing climate, and improve the resilience of regional water resources. The relative proportions and roles map of climate change and human activities in vegetation change areas provide a basis for government to formulate local-based management policies.

How to cite: Xue, L., Kappas, M., Wyss, D., and Putzenlechner, B.: Assessment of climate change and human activities on vegetation development in Northeast China, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-7718, https://doi.org/10.5194/egusphere-egu22-7718, 2022.

Josep Roca and Blanca Arellano

There is no universal definition of a heatwave, but extreme events associated with particularly hot, sustained temperatures have been known to have a notable impact on human mortality, regional economies and ecosystems. In this paper, we use the concept of heatwave applied by the Spanish Meteorological Agency (AEMET). In this definition, a heatwave is considered an episode of at least three consecutive days in which the stations that are considered register maximums above the 95% percentile of the series of maximum daily temperatures for the months of July and August from the period 1971 to 2000. However, this definition has a major limitation: it refers only to maximum temperatures, not minimum ones. Maximum temperatures can have serious consequences, especially on heat stroke. However, the health effects are more pronounced in the case of night heat, where the inability to rest (especially in homes without air conditioning, as is generally the case in in Spain) can cause significant worsening of respiratory and cardiovascular diseases that produce a high proportion of premature deaths. For this reason, in this study we differentiate between heatwaves during the day (DHW) and at night (NHW), paying special attention to the latter.

The research aims to study extreme heat events in the city of Barcelona between 1971 and 2020. Since the urban climate presents a marked spatial variation, taking into account the geographical characteristics of the territory, as well as the spatial distribution of the island of urban heat, the research is carried out based on the information provided by four representative meteorological stations of the study area: Fabra Observatory, CMT, Raval and Barcelona Airport.

The maximum average temperatures at the Fabra Observatory, and for the last 50 years, increased 2.88 degrees, which represents an annual increase of 0.058 degrees/year. The minimum average temperature increased 2.38 degrees, 0.048 degrees/year. However, the increases differ very significantly depending on the spatial location of the meteorological station. The results are quite different for other observatories, as Barcelona airport. At the airport, the increase in maximum temperature was less prominent (2 °C in the last 50 years, 10.3%). However, the minimum temperatures increased by 35.8%; 3.62 °C between 1971 and 2020. An OLS model, with the maximum and minimum daily temperatures of the last 50 years from various stations (Fabra, Airport, Raval and CMT), and using the year, the month and the calendar day (cd *) as explanatory variables, generally confirmed the warming process in the Barcelona area.

Therefore, global warming is a clear reality in the Mediterranean area in which Barcelona is located. The work shows a marked difference in extreme heat events between different urban locations. The proximity to the sea, the altitude, the different urban density and the quantity and quality of urban greenery have a determining effect on daytime and nighttime heat waves.

How to cite: Roca, J. and Arellano, B.: Day and night heat waves in the city of Barcelona. 1971-2020., EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-6909, https://doi.org/10.5194/egusphere-egu22-6909, 2022.

Q&A Human influence and climatic trends. Closing of second block