Search

In climate science and applications, the term "metric" is used to describe the distillation of complex, multifaceted evaluations to summarize the overall quality of a model simulation, or other data product, and/or as a means to quantify some response to climate change. Metrics provide insights into the fidelity of processes and outcomes from climate models and can assist with both differentiating models' representation of variables or processes and informing whether models are "fit for purpose." Metrics can also provide a valuable reference point for co‐production of knowledge between climate scientists and climate impact practitioners. Although continued metric developments enable model developers to better understand the impacts of decisions made in the model design process, metrics also have implications for the characterization of uncertainty and facilitating analyses of underlying physical processes. As a result, comprehensive evaluation with multiple metrics enhances usability of climate information by both scientific and stakeholder communities. This paper presents examples of insights gained from the development and appropriate use of metrics, and provides examples of how metrics can be used to engage with stakeholders and inform decision‐making. This article is categorized under:Climate Models and Modeling > Knowledge Generation with ModelsThe Social Status of Climate Change Knowledge > Climate Science and Decision MakingAssessing Impacts of Climate Change > Evaluating Future Impacts of Climate Change [ABSTRACT FROM AUTHOR]

With the advent of climate change as a major challenge of our time, Earth system modeling has become highly policy‐relevant regulatory science. In this situation, the social mechanisms that play a role in any scientific endeavor become particularly exposed. By discussing historical, philosophical, and sociological (HPS) aspects of the field's current "cultures of prediction" together with the physical science community in a physical science journal, we aim to provide an entry point into HPS reasoning for climate scientists interested in reflecting on their field and science in general. This paper, first, introduces our perspective on "science as culture" and climate modeling as "regulatory science" and, second, highlights and connects relevant ideas from the three commentaries that follow it. In so doing, we hope to give a fuller picture of climate science, the interplay it engenders between HPS and the physical sciences, the distinctions that it gives rise to as compared to some of the more traditional, exact, sciences in which it is rooted and its place in society including its role in scientific policy advice. Key Points: With the advent of climate change as a major challenge of our time, Earth system modeling has become highly policy relevantIn this situation, the social mechanisms that play a role in any scientific endeavor become particularly exposedThis paper introduces perspectives on "science as culture" and highlights relevant ideas from the three subsequent commentaries [ABSTRACT FROM AUTHOR]

Vb cyclones typically emerge in the Western Mediterranean and propagate to the Northeast into Central Europe. This paper explores the temporal characteristics of Vb cyclone occurrence based on cyclone tracks identified at the atmospheric levels of Z700 and sea level pressure, using JRA‐55 reanalysis data for the period 1959–2015. The risk of Vb occurrence was significantly high in the 1960s and has remained at a lower level since then. Vb cyclones do not occur fully randomly according to a Poisson point process. Eleven well‐separated and distinct clusters as well as 11 hiatus periods are identified, with average occurrence rates of 21.5 and 5.2 yrea−1, respectively. During the event of Vb, the large‐scale atmospheric circulation is changed into a state favoring the development of successive Vb cyclones. Clustering is very prominent in the case of Genoan Vb cyclones in summer as well as those Vb cyclones developing over the Iberian Peninsula or the North African Coast in winter. Superposition of the polar and the subtropical jet stream over the Western Mediterranean is identified as a main feature at the onset of Vb cyclones. Vb cyclone occurrence appears to be synchronized with the Northern Atlantic Oscillation (NAO; at Z500) and Arctic Oscillation (AO; at Z1000). Clusters have occurred when both NAO and AO were negative. This relation applies to Western Mediterranean cyclones not following a Vb track as well, however to a much weaker extent. In contrast, Vb cyclone frequency was particularly low from 1988 to 1997 during a sustained positive phase of both NAO and AO. Key Points: In the past 50 years, the frequency of Vb cyclones was high in the 1960s and lower since thenVb cyclones were strongly synchronized with NAO and AO; high frequency clusters occurred when NAO and AO were negativeThe coupling of the polar and the subtropical jet stream over the Western Mediterranean is a driving mechanism of the onset of Vb cyclones [ABSTRACT FROM AUTHOR]

This paper compares extreme value statistics of daily maximum 10 m wind speed in winter simulated by the regional climate model COSMO‐CLM at a horizontal resolution of 15 km (C15) with the reanalyses ERA‐Interim and Arctic System reanalysis (ASR version 1 and 2) and with a satellite data set (CCMPv2). Our C15 simulation (1979/1980–2015/2016, November–April) is thereby the longest high‐resolution simulation available for the Arctic. The results show that the extreme wind speeds tend to increase over the ocean with increasing the horizontal model resolution. A horizontal resolution of ≤15 km is required to sufficiently capture all extreme wind characteristics, in particular for the tip jets and barrier winds over the Irminger Sea and in the Denmark Strait, and for the low‐level jets in the Nares Strait. Of almost equal importance are physical parameterizations of surface fluxes and of turbulence. Capturing extreme wind characteristics has a direct effect on climate relevant air–ice–ocean interactions, such as triggering open‐ocean deep convection, polynya dynamics, or on the sea ice and freshwater balance of the Arctic. State‐of‐the‐art global climate models and reanalyses (e.g., ERA‐Interim) underestimate the extreme near‐surface wind speeds in the Arctic and around Greenland. Regional climate models, for example, ASR/WRF or COSMO‐CLM, resolve the mesoscale wind systems that frequently produce extreme winds, which we estimated i.a. as 10‐year return levels (see the figure). Capturing extreme wind characteristics has a direct effect on climate relevant air–ice–ocean interactions, such as triggering open‐ocean deep convection, polynya dynamics, or on the sea ice and freshwater balance of the Arctic. [ABSTRACT FROM AUTHOR]

To better understand the regional changes in summertime temperatures across the conterminous United States (CONUS), we adopt a recently developed machine learning framework that can be used to reveal the timing of emergence of forced climate signals from the noise of internal climate variability. Specifically, we train an artificial neural network (ANN) on seasonally averaged temperatures across the CONUS and then task the ANN to output the year associated with an individual map. In order to correctly identify the year, the ANN must therefore learn time‐evolving patterns of climate change amidst the noise of internal climate variability. The ANNs are first trained and tested on data from large ensembles and then evaluated using observations from a station‐based data set. To understand how the ANN is making its predictions, we leverage a collection of ad hoc feature attribution methods from explainable artificial intelligence (XAI). We find that anthropogenic signals in seasonal mean minimum temperature have emerged by the early 2000s for the CONUS, which occurred earliest in the Eastern United States. While our observational timing of emergence estimates are not as sensitive to the spatial resolution of the training data, we find a notable improvement in ANN skill using a higher resolution climate model, especially for its early twentieth century predictions. Composites of XAI maps reveal that this improvement is linked to temperatures around higher topography. We find that increases in spatial resolution of the ANN training data may yield benefits for machine learning applications in climate science. Plain Language Summary: While temperatures around the world continue to warm due to human‐caused climate change, some areas have observed smaller temperature trends than others. Understanding this regional variability in the rate of warming is important when assessing future projections. One location that has observed less warming is across the United States during their summer season. To evaluate temperature variability in this region using real‐world observations and climate model simulations, we use a statistical method from artificial intelligence called neural networks. The goal of the neural network setup is to learn temperature patterns across the United States and then identify whether climate change effects have exceeded the range of natural variability that has occurred in the past. This is called the timing of emergence (ToE), which is the first year that the effect has clearly appeared. We find that the average United States minimum temperature increase has already emerged in historical records. However, we find no ToE for the average maximum temperature, other than in the Western United States. Another important finding of this study is that by using higher resolution climate model data (i.e., more latitude and longitude points), we find better accuracy in the neural network predictions. Key Points: Forced temperature changes have emerged in observations during summer in the United States as detected by an artificial neural networkIncreasing spatial resolution improves neural network skill for predicting the year of a given summer temperature mapWestern United States land surface climate properties contribute to earlier timing of emergence predictions for the SPEAR climate model [ABSTRACT FROM AUTHOR]

ABSTRACT The aim of this paper is to study the model's simulated frequency, track, intensity and location of cyclonic storms (CSs) and severe cyclonic storms (SCSs) in the Bay of Bengal (BoB), India. For the purpose, the PRECIS (Providing REgional Climates for Impacts Studies) a regional climate modelling system, of UK Met Office, is used. This model is integrated for the period 1961-1990 (baseline) and the future time period 2071-2100 (High emission scenario, A2). To run the model, the initial and lateral boundary conditions are supplied by UK Met Office. The analysis of frequency, track, intensity and location are carried out for May, June, September and October for the period 1961-1990 and 2071-2100. To evaluate the model's performance in simulating storms frequency during 1961-1990, chi square test is carried out with observed storms for the same period. The model's simulated frequency of storms is an overestimation of observations although the frequency of model's simulated storms during 2071-2100 is less than that of during 1961-1990. In general, model's simulated storms are found moving in the northwest direction from their initial location in all months during 1961-1990 and 2071-2100. In model's simulations, the drop in central pressure is relatively more in the months of May, June, September and October during 2071-2100 compared with that during 1961-1990. During 2071-2100, more intense storms may be possible in the months of May and June compared with that of September and October. It is observed that the model is able to simulate the initial locations of storms during 1961-1990 and 2071-2100 close to observations for the months of June and September, especially. [ABSTRACT FROM AUTHOR]

The increasing awareness of climate change requires accurate, reliable and timely information on possible precipitation (PRE) changes to build climate resilience. This study uses the Coupled Model Intercomparison Project phase 6 (CMIP6) data to examine the effectiveness of bias correction on simulated historical mean and extreme PRE, and investigates the projected changes in extreme PRE events over southern Africa (SAF). Quantile mapping on a gamma distribution bias correction method is applied to the historical and projected CMIP6 data, with the Global Precipitation Climatology Centre (GPCC) PRE dataset as reference data. The projection analyses are conducted for the wet (December–March) and dry (May–October) seasons for two shared socioeconomic pathway (SSP) scenarios: SSP2‐4.5 and SSP5‐8.5. The results affirm that the bias correction significantly reduces (by more than 90%) the biases in the modelled mean and extreme PRE. The projected extreme PRE shows a general drying pattern over SAF. In the wet season, wet days (R1mm) are projected to decrease by 1 and 2 days by the end of the 21st century for SSP2‐4.5 and SSP5‐8.5, respectively, while very heavy PRE days (R20mm) show a slight upward pattern over SAF, reaching 0.1 and 0.4 days for the two scenarios. In contrast, the dry season exhibits a more pronounced drying tendency, with a consistent decrease in R1mm of 2 and 0.1 days by 2100 for SSP2‐4.5 and SSP5‐8.5, respectively. R20mm is projected to decrease by 4 and 0.2 days for SSP2‐4.5 and SSP5‐8.5, respectively. This implies that SAF is expected to get drier during the dry season while the PRE intensity during the wet season is projected to increase. [ABSTRACT FROM AUTHOR]

The ability of regional climate models (RCMs) to accurately simulate the current climate is increasingly important for impact assessments over Southeast Asia (SEA), identified as one of the world's most vulnerable regions to climate change. In this study, we evaluate the performance of a set of regional high‐resolution simulations from the Coordinated Regional Climate Downscaling Experiment‐SEA (CORDEX‐SEA) in simulating rainfall over the region. Simulations of the 1982–2005 seasonal mean climatology of daily precipitation and precipitation distribution over land are compared to observations from different sources (i.e., in situ‐based and satellite‐based). We also evaluate to what extent the precipitation distribution in RCMs is closer to observations than their associated forcing global climate models (GCMs). Observational estimates of precipitation over SEA have large uncertainties, making the model evaluations complicated. Despite these difficulties, our results highlight that RCMs can reproduce some complexities in the spatial distribution of seasonal rainfall but generally have a larger wet bias than GCMs. This is particularly true for the extremes in which RCMs show a large overestimation of rainfall intensity. There are some precipitation quantiles and grid points in which RCMs show limited reductions in biases compared to observations, but there is no consistency across all simulations and RCMs are generally further away from observations than their forcing GCMs. We find that greater intensity in RCMs over CORDEX‐SEA compared to their associated forcing GCMs is firstly associated with the increased supply of moisture from both local and large‐scale sources. Second, a widespread increase in convective precipitation is found across the region in RCMs. Our findings suggest that a model's ability to simulate precipitation over the region relies more on the RCM setup itself (e.g., parameterization scheme), rather than its forcing GCM. This should be considered when assessing the reliability of RCM precipitation simulations for future projections. [ABSTRACT FROM AUTHOR]

The global climate models (GCMs) performances of the recently released Coupled Model Intercomparison Project phase 6 (CMIP6) compared to its predecessor, CMIP5, are evaluated to anticipate the expected changes in climate over Egypt. Thirteen GCMs and their multi‐model ensemble (MME) of both CMIPs were used for this purpose. The future projections were compared for two radiative concentration pathways (RCP 4.5 and 8.5) and two shared socio‐economic pathways (SSP 2–4.5 and 5–8.5). The results revealed improvement in most CMIP6 models in replicating historical rainfall, maximum temperature (Tmax), and minimum temperature (Tmin) climatology over Egypt. The MME of the CMIPs revealed that both could reproduce Egypt's spatial distribution and seasonal climate variability. However, the bias in CMIP5 was higher than that for CMIP6. The uncertainties in simulating seasonal variability of rainfall and temperatures were lower for CMIP6 compared to CMIP5. The future projection of rainfall using CMIP6 MME revealed a higher reduction of precipitation (10–26 mm) in the economically crucial northern region than that estimated using CMIP5 (0–17 mm), compared to 133.5 mm rainfall in the base period. CMIP6 also projected a 0.74–1.63°C more rise in Tmax and Tmin compared to CMIP5 by the end of the century. The study indicates more aggravated scenarios of climate changes in Egypt than anticipated earlier, using the CMIP5 models. Therefore, Egypt needs to streamline the existing adaptation and mitigation measures to account for climate projections. [ABSTRACT FROM AUTHOR]

This study evaluated the rainfall historical simulations of 15 Global Climate Models (GCMs) of the Coupled Model Intercomparison Project phase 6 (CMIP6) in replicating annual and seasonal rainfall climatology, their temporal variability and trends in Bangladesh for the period 1979–2014, considering ERA5 (ECMWF Reanalysis 5th Generation) reanalysis as the reference dataset. Shannon's Entropy decision‐analysis was employed for GCMs' rating based on eight statistical indicators and a comprehensive rating metric for the final grading of the GCMs. The majority of the CMIP6 GCMs accurately reproduced the spatial feature of ERA5 rainfall. However, the GCMs underestimated annual rainfall by an average of 190.5 mm, with the highest underestimation in monsoon (131.76 mm) and least in winter (3.52 mm) seasons. Most GCMs also underestimated rainfall variability for all seasons except winter. Besides, the GCMs showed an increasing trend in pre‐monsoon and a decreasing trend in post‐monsoon rainfall like ERA5, but an opposite (negative) to ERA5 trend (positive) in monsoon season rainfall. The ensemble mean of the GCMs showed higher skill in reconstructing rainfall climatology, temporal variability and trends than the individual GCMs. The study identified MPI‐ESM1‐2‐LR, MPI‐ESM1‐2‐HR, and GFDL‐ESM4 as the most effective GCMs in reproducing precipitation over Bangladesh. The selected models' simulation can be used for climate change impact assessment in Bangladesh after bias minimization. [ABSTRACT FROM AUTHOR]

Despite repeated calls for greater collaboration between physical and human geographers, the unique interdisciplinary potential of geography remains largely underutilized. Yet geographers are well positioned to take a leading role in the interdisciplinary turn in climate‐related research. This paper explores the possibilities for physical and human geographers to collaborate within and beyond the discipline, specifically on the topic of climate modeling. We first examine geographical research critically examining the production and circulation of climate knowledge. Drawing on insights from a recent literature called "Critical Physical Geography," we then outline how geographers might engage in collaborative and interdisciplinary work in order to promote more democratic practices of producing climate knowledge, enrich understandings of climate change, and more effectively serve goals of social and environmental justice. We argue that both the discipline of geography and the field of climate research stand to gain enormously from geographers' efforts to talk across the divide between social and natural science within and beyond the boundaries of geography. [ABSTRACT FROM AUTHOR]

The persistent inter‐model spread in the response of global‐mean surface temperature to increased CO2 (known as the "Equilibrium Climate Sensitivity," or "ECS") is a crucial problem across model generations. This work examines the influence of the models' present‐day atmospheric circulation climatologies, and the accompanying climatological cloud radiative effects, in explaining that spread. We analyze the Coupled Model Intercomparison Project Phase 6 (CMIP6) models and find that they simulate a more poleward, and thus more realistic, edge of the Hadley cell (HC) in the Southern Hemisphere than the CMIP5 models, although the climatological shortwave cloud radiative effects are similar in the two generations of models. A few CMIP5 models with extreme equatorward biases in the HC edge exhibited high ECS due to strong Southern midlatitude shortwave cloud radiative warming in response to climate change, suggesting an ECS dependence on HC position. We find that such constraint no longer holds for the CMIP6 models, due to the absence of models with extreme HC climatologies. In spite of this, however, the CMIP6 models show an increased spread in ECS, with more models in the high ECS range. In addition, an improved representation of the climatological jet dynamics does not lead to a new emergent constraint in the CMIP6 models either. [ABSTRACT FROM AUTHOR]

This study, for the first time, investigates the physical association between Predecessor Rain Events (PREs) and peak runoff generation in seven catchments over the Upper Mahanadi River basin (UMRB), India. A statistical–dynamical framework is developed to assess the compounding impact of PREs (as preconditioned events) versus riverine floods during both retrospective and projected climate. Based on models' fidelity to capture historical climatology, we select four out of nine Global Climate Models (GCMs) during historical (1980–2005), and the three projected time windows, that is, near future (2010–2039), mid‐century (2040–2069), and the far future (2070–2099) planning horizons with RCP8.5 emission scenario. We assess changes in compound flood hazards in historical versus projected periods using a newly proposed Bivariate Hazard Ratio (BHR) index, which represents the ratio between bivariate return periods during the projected versus the historical time windows. Assessing bivariate return periods (characterized by 'AND' operator) of rainfall‐driven compound floods shows decreased flood hazard in the mid‐century and far‐future planning horizons. Accounting ranges of uncertainty from climate model simulations and the propagation of uncertainty across the numerical model chain, overall we show considering PRE as the covariate, floods in larger catchments show an increase in compound flood hazard in the projected period. [ABSTRACT FROM AUTHOR]

Natural climate variability poses a limit to the confidence in regional climate projections, particularly for the near‐term. Therefore, the next decades are of importance for regional impact studies and the development of regional adaptation plans. This article presents results from an ensemble of regional climate model simulations (years 1996–2005 and 2026–2035) that are used to examine the effects of both anthropogenic forcing and natural variability associated with the Pacific Decadal Oscillation (PDO) on near‐term climate projections for the Hawaiian Islands. The Community Earth System Model Large Ensemble (CESM‐LE) is used in conjunction with the Weather Research and Forecasting (WRF) model for downscaling. The climate responses to the PDO and anthropogenic forcing are isolated and analyzed separately. In response to anthropogenic forcing, significant increases in surface air temperature, of ∼0.8 K, are projected at low elevations. Stronger warming, of up to 1.3 K, is seen at higher elevation areas. Future climate simulations show significant increases in wet season rainfall, of ∼10%–20%, along the windward slopes of Big Island and Maui. Rainfall patterns during the positive PDO phase are projected to reverse in sign, leading to drier conditions, by ∼10%–30%, at many locations. Future climate simulations show daily rainfall extremes will increase, by up to ∼10%–15%, at many locations. Daily temperature extremes are also projected to increase significantly, by up to 1.4 K. Overall, results indicate that natural variability will continue to contribute to uncertainty in near‐term rainfall projections for the Hawaiian Islands, masking the forced signal. Plain Language Summary: This project studies how temperature and precipitation will change across the Hawaiian Islands in the near future due to human‐induced causes (e.g., burning coal and other fossil fuels). This is done by comparing climate model simulations for two 10‐year periods (1996–2005 vs. 2026–2035). Further, this research evaluates how natural climate variations affect Hawaiian climate. The main analysis concentrates on an important type of natural climate variability in the North Pacific, the Pacific Decadal Oscillation. This research implements dynamical downscaling, that is, a regional climate model is used to refine global climate model simulations in order to better capture the local weather patterns in Hawai'i. We find that land and ocean temperatures across the Hawaiian Islands region will become warmer due to human‐induced causes. We also find that rainfall will likely increase for some areas of Hawai'i, particularly the mountain slopes, posing an increased risk for flooding. The simulations reveal that natural climate variability will continue to contribute to near‐term climate variations in this region. While the natural climate variability limits our confidence in the projected rainfall changes, it is hoped that our findings will assist decision makers across the State of Hawai'i in creating climate adaptation plans. Key Points: Significant increases in near‐surface temperature and sea surface temperature are projected across the Hawaiian Islands regionSignificant increases in mean and extreme rainfall are projected for some areas of the islandsNatural decadal variability will continue to mask the forced signal in near‐term rainfall projections for the Hawaiian Islands [ABSTRACT FROM AUTHOR]

The use of regional climate model (RCM)‐based projections for providing regional climate information in a research and climate service contexts is currently expanding very fast. This has been possible thanks to a considerable effort in developing comprehensive ensembles of RCM projections, especially for Europe, in the EURO‐CORDEX community (Jacob et al., 2014, 2020). As of end of 2019, EURO‐CORDEX has developed a set of 55 historical and scenario projections (RCP8.5) using 8 driving global climate models (GCMs) and 11 RCMs. This article presents the ensemble including its design. We target the analysis to better characterize the quality of the RCMs by providing an evaluation of these RCM simulations over a number of classical climate variables and extreme and impact‐oriented indices for the period 1981–2010. For the main variables, the model simulations generally agree with observations and reanalyses. However, several systematic biases are found as well, with shared responsibilities among RCMs and GCMs: Simulations are overall too cold, too wet, and too windy compared to available observations or reanalyses. Some simulations show strong systematic biases on temperature, others on precipitation or dynamical variables, but none of the models/simulations can be defined as the best or the worst on all criteria. The article aims at supporting a proper use of these simulations within a climate services context. Plain Language Summary: This study analyses the ability of the unprecedently large ensemble of 55 regional climate simulations to properly simulate the climatology of several variables, extremes, and impact‐oriented indices over the European continent. This analysis should guide the use of regional climate projections in climate services development. Key Points: Biases of an unprecedentedly large ensemble of 55 European climate simulations using 8 global climate models and 11 regional climate models are assessedClimate variables, extremes, and impact‐oriented indices are assessed, indicating whether such ensemble can—or cannot—be used in climate service applicationsSimulations are generally too wet, too cold and too windy, and the share of contributions to the bias from GCMs and RCMs is found to differ for each variable or index [ABSTRACT FROM AUTHOR]

Given uncertainty in the processes involved in polar amplification, elucidating the role of poleward heat and moisture transport is crucial. The Polar Amplification Model Intercomparison Project (PAMIP) permits robust separation of the effects of sea ice loss from sea surface warming under climate change. We utilize a moist isentropic circulation framework that accounts for moisture transport, condensation, and eddy transport, in order to analyze the circulation connecting the mid‐latitudes and the Arctic. In PAMIP's atmospheric general circulation model experiments, prescribed sea ice loss reduces poleward heat transport (PHT) by warming the returning moist isentropic circulation at high latitudes, while prescribed warming of the ocean surface increases PHT by strengthening the moist isentropic circulation. Inter‐model spread of PHT into the Arctic reflects the tug‐of‐war between sea‐ice and surface‐warming effects. Plain Language Summary: A major conundrum in current climate science is to understand what Arctic changes imply for the climate and environment in mid‐latitude regions. The Polar Amplification Model Intercomparison Project (PAMIP) designed a set of climate model experiments to specifically answer this question in a carefully designed, idealized framework. PAMIP's approach is to separate historic and projected climate change into parts associated with Arctic sea ice loss and ocean surface warming and investigate how these two contributions can influence the atmosphere. To isolate these effects, only models can be used, because, in reality, sea ice loss and ocean surface warming are strongly linked together. This letter focuses on what the PAMIP experiments imply for the transport and redistribution of heat and moisture in the atmosphere. In PAMIP, we learn that Arctic sea ice loss causes the atmosphere to reduce the transport of dry and cold air away from the Arctic while ocean warming causes more transport of moist warm air toward the Arctic. These two effects are in a tug‐of‐war, suggesting that climate change can cause a mix of competing impacts on the global energy transport from the tropics to the Arctic. Key Points: Sea ice loss exports static energy from the Arctic by warming the equatorward branch of the global mass circulationSea surface warming imports static energy into the Arctic by increasing the mid‐latitude mass transport, overwhelming the sea ice effectThere is more spread amongst models for the effect of sea‐surface warming than for sea ice loss [ABSTRACT FROM AUTHOR]

Risks associated to extreme hydrological conditions, such as floods and droughts, are expected to increase in future climate because of projected changes in precipitation and temperature. Assessing how wet and dry persisting conditions (or spells) will evolve in future climate is a crucial step in the study of extreme hydrological events. Projected changes over the north‐eastern part of North America in the annual number of wet days, wet, and dry spells characteristics (number, duration), and wet spells intensities are analyzed. Two regional climate model ensembles are considered: the multimember ensemble from the Canadian RCM v5 (CRCM5‐LE), and 16 simulations from the NA‐CORDEX multimodel ensemble both using the Representative Concentration Pathway 8.5 (RCP8.5) scenario. Comparison of observed and simulated wet and dry spells characteristics is first performed in present climate (1971–2000). Regional Climate Model (RCM)s tend to generate too many wet days and wet or dry spells resulting in shorter dry spells and slightly longer wet spells. Modeling uncertainties are accounted for a bigger contribution to the bias than internal variability since the multimodel ensemble dispersion is the largest. Throughout the 21st century, both ensembles project significant trends in winter at higher latitudes resulting in increasing wet day frequency, increasing number of wet spell, longer wet spells and shorter dry spells. For other seasons, internal variability of the CRCM5‐LE and differences among the various NA‐CORDEX simulations seems to overwhelm the climate change signal. Wet spell intensities increases are projected for all seasons over almost the entire domain. Globally, wetter climate with potential significant hydrological impacts are expected in many regions. Key Points: Wet and dry spells future evolution are assessed using two RCMs ensemble: CRCM5‐LE and NA‐CORDEXModeling uncertainties are accounted for a bigger contribution to the bias than internal variabilityBoth internal variability and modeling uncertainties seems to overwhelm the climate change signal [ABSTRACT FROM AUTHOR]

The climate science community is challenged to adopt an actionable risk perspective, which is difficult to align with the traditional focus on model‐based probabilistic climate change projections. Event‐based storylines can provide a way out of this conundrum by putting emphasis on plausibility rather than probability. This links directly to common practices in disaster risk management using "stress‐testing" for emergency preparedness based on events that are conditional on specific and plausible assumptions. Event‐based storylines allow for conditional explanations, without full attribution of every causal factor, which is crucial when some aspects of the latter are complex and highly uncertain. Plain Language Summary: One of today's major challenges is how to use insights and information from climate sciences to inform decision‐making regarding managing risks from climate change, where weather and climate extremes represent a major component of climate‐related risk. So far, climate science has taken a probabilistic approach producing large model ensembles and exploring likely ranges, thereby neglecting low‐likelihood but potentially high‐impact events that pose significant risks to society. Event‐based storylines are emerging as an alternative way to explore future high‐impact events while taking into account aspects of vulnerability and exposure of the considered system with an emphasis on plausibility rather than probability. This concept links directly to common practices in disaster risk management using "stress‐testing" for emergency preparedness based on events that are conditional on specific, but plausible assumptions. When co‐developed by climate scientists and stakeholders, event‐based storylines can be informed by physical climate and impact modeling and can provide a useful way of communicating and assessing climate‐related risk in a specific decision‐making context. Key Points: Event‐based storylines are a way to communicate and assess climate risk taking into account aspects of vulnerability and exposureEvent‐based storylines focus on plausibility rather than probability when looking at high‐impact eventsEvent‐based storylines can provide climate information that feeds directly into a particular decision‐making context [ABSTRACT FROM AUTHOR]

The purpose of this study is to assess the ability of a set of large ensembles (LE) of Global Climate Model (GCM) simulations from the Multi‐Model Large Ensemble Archive, Coupled Model Intercomparison Project Phase 6 (CMIP6) and database for Policy Decision making for Future climate change in reproducing the variability and change of the austral summer precipitation observed in South America along the second part of the 20th century and beginning of the 21st. LE show similar biases in the mean austral summer rainfall and interannual variability than those detected in previous model sets, such as the Coupled Model Intercomparison Project Phase 5 (CMIP5). The positive trends in south‐eastern South America (SESA) and negative ones in the southern Andes (SAn), that are the most significant observed regional features, are identified to some extent in LE simulations. The negative trend in SAn is a feature consistently shown among different models. For all models, mean trend is negative in that particular region and larger than the inter‐member dispersion for each large ensemble. Trend magnitudes in the SESA region show a larger dispersion between models than the SAn region. While the mean trend among models is consistent with the observations, models underestimate the observed trend. The multi‐model ensemble of selected models that best reproduce both mean and interannual variability features of rainfall in South America shows an 18% larger positive mean trend in SESA rainfall than those that have the worst performance. However, internal variability uncertainty is still higher than the mean trend for both sets of multi‐model ensembles [5% (18%) for selected (non‐selected) models]. Improvements in the representation of the main features of mean rainfall and variability are needed in order to better describe and reduce the uncertainties associated with the main climate change related signals in the continent. [ABSTRACT FROM AUTHOR]

Climate impact studies often require climate data at a higher space–time resolution than is available from global and regional climate models. Weather generator (WG) models, generally designed for mesoscale applications (e.g., 101–105 km2), are popular and widely used tools to downscale climate data to finer resolution. One advantage of using WGs is their ability to generate the necessary climate variables for impact studies in data sparse regions. In this study, we evaluate the ability of a previously established state of the art WG (the AWE‐GEN‐2d model) to perform in data sparse regions that are beyond the mesoscale, using the Zambezi River basin (106 km2) in southeast Africa as a case study. The AWE‐GEN‐2d model was calibrated using data from satellite retrievals and climate re‐analysis products in place of the absent observational data. An 8‐km climate ensemble at hourly resolution, covering the period of 1976–2099 (present climate and RCP4.5 emission scenario from 2020), was then simulated. Using the simulated 30‐member ensemble, climate indices for both present and future climates were computed. The high‐resolution climate indices allow detailed analysis of the effects of climate change on different areas within the basin. For example, the southwestern area of the basin is predicted to experience the greatest change due to increased temperature, while the southeastern area was found to be already so hot that is less affected (e.g., the number of 'very hot days' per year increase by 18 and 9 days, respectively). Rainfall intensities are found to increase most in the eastern areas of the basin (1 mm·d−1) in comparison to the western region (0.3 mm·d−1). As demonstrated in this study, AWE‐GEN‐2d can be calibrated successfully using data from climate reanalysis products in the absence of ground station data and can be applied at larger scales than the mesoscale. [ABSTRACT FROM AUTHOR]

In climate science, observational gridded climate datasets that are based on in situ measurements serve as evidence for scientific claims and they are used to both calibrate and evaluate models. However, datasets only represent selected aspects of the real world, so when they are used for a specific purpose they can be a source of uncertainty. Here, we present a framework for understanding this uncertainty of observational datasets which distinguishes three general sources of uncertainty: (1) uncertainty that arises during the generation of the dataset; (2) uncertainty due to biased samples; and (3) uncertainty that arises due to the choice of abstract properties, such as resolution and metric. Based on this framework, we identify four different types of dataset ensembles—parametric, structural, resampling, and property ensembles—as tools to understand and assess uncertainties arising from the use of datasets for a specific purpose. We advocate for a more systematic generation of dataset ensembles by using these sorts of tools. Finally, we discuss the use of dataset ensembles in climate model evaluation. We argue that a more systematic understanding and assessment of dataset uncertainty is needed to allow for a more reliable uncertainty assessment in the context of model evaluation. The more systematic use of such a framework would be beneficial for both scientific reasoning and scientific policy advice based on climate datasets. This article is categorized under: Paleoclimates and Current Trends > Modern Climate Change [ABSTRACT FROM AUTHOR]

Global warming acts to prolong thermal summers and shorten winters. In this work, future changes in the lengths and timing of four thermal seasons in northern Europe, with threshold temperatures 0 and 10°C, are derived from bias‐adjusted output data from 23 CMIP5 global climate models. Three future periods and two Representative Concentration Pathway (RCP) scenarios are discussed. The focus is on the period 2040–2069 under RCP4.5, which approximately corresponds to a 2°C global warming relative to the preindustrial era. By the period 2040–2069, the average length of the thermal summer increases by nearly 30 days relative to 1971–2000, and the thermal winter shortens by 30–60 days. The timing of the thermal springs advances while autumns delay. Within the model ensemble, there is a high linear correlation between the modelled annual‐mean temperature increase and shifts in the thermal seasons. Thermal summers lengthen by about 10 days and winters shorten by 10–24 days per 1°C of local warming. In the mid‐21st century, about two‐thirds of all summers (winters) are projected to be very long (very short) according to the baseline‐period standards, with an anomaly greater than 20 days relative to the late‐20th century temporal mean. The proportion of years without a thermal winter increases remarkably in the Baltic countries and southern Scandinavian peninsula. Implications of the changing thermal seasons on nature and human society are discussed in a literature review. [ABSTRACT FROM AUTHOR]

Variable‐resolution global climate models (VRGCMs) are a dynamical downscaling method that can reach spatiotemporal scales needed for regional climate assessments. Over the years, several users of VRGCMs have assumed where the location and extent of the refinement domain should be based on knowledge of the prevailing storm tracks and resolution dependence of important regional climate processes (e.g., atmospheric rivers [ARs] and orographic uplift), but the effect of high‐resolution domain size and extent on the simulation of downstream hydroclimatic phenomena has not been systematically evaluated. Here, we use variable resolution in the Community Earth System Model (VR‐CESM) to perform such a test. To do this, three VR‐CESM grids were generated that span the entire, two thirds, and one third of the North Pacific and evaluated for a 30‐year climatology using Atmospheric Model Intercomparison Project protocols. Simulations are compared with reanalysis products offshore (fifth‐generation of the European Centre for Medium‐Range Weather Forecasts atmospheric reanalysis [ERA5]) and onshore (Livneh, 2015, https://doi.org/10.1038/sdata.2015.42, and Parameter‐elevation Regressions on Independent Slopes Model [PRISM]) of the western United States. The westward expansion of refinement domain influenced integrated vapor transport (IVT), which was generally high biased but minimally impacted AR characteristics. Due to slight differences in landfalling AR counts in the western United States, California winter precipitation generally improved with westward expansion of the refinement domain. Western U.S. mountain snowpack and surface temperatures were insensitive to refinement domain size and were more influenced by changes in topographic resolution and/or land surface model version. Given minimal dependence of simulated western U.S. hydroclimate on refinement domain size over the North Pacific, we advise future VR‐CESM studies to focus grid resolution on better resolving land surface heterogeneity. Key Points: Three variable‐resolution Community Earth System Model simulations are assessed for sensitivities to refinement domain sizeMore extensive refinement of the western Pacific reduces the integrated water vapor transport bias to the western United StatesTopographic resolution and land surface model have greater influences on simulated hydroclimatology than refinement domain extent [ABSTRACT FROM AUTHOR]

Outputs from new state‐of‐the‐art climate models under the Coupled Model Inter‐comparison Project phase 6 (CMIP6) promise improvement and enhancement of climate change projections information for Australia. Here we focus on three key aspects of CMIP6: what is new in these models, how the available CMIP6 models evaluate compared to CMIP5, and their projections of the future Australian climate compared to CMIP5 focussing on the highest emissions scenario. The CMIP6 ensemble has several new features of relevance to policymakers and others, for example, the integrated matrix of socioeconomic and concentration pathways. The CMIP6 models show incremental improvements in the simulation of the climate in the Australian region, including a reduced equatorial Pacific cold tongue bias, slightly improved rainfall teleconnections with large‐scale climate drivers, improved representation of atmosphere and ocean extreme heat events, as well as dynamic sea level. However, important regional biases remain, evident in the excessive rainfall over the Maritime Continent and rainfall pattern biases in the nearby tropical convergence zones. Projections of Australian temperature and rainfall from the available CMIP6 ensemble broadly agree with those from CMIP5, except for a group of CMIP6 models with higher climate sensitivity and greater warming and increase in some extremes after 2050. CMIP6 rainfall projections are similar to CMIP5, but the ensemble examined has a narrower range of rainfall change in austral summer in Northern Australia and austral winter in Southern Australia. Overall, future national projections are likely to be similar to previous versions but perhaps with some areas of improved confidence and clarity. Key Points: CMIP6 shows incremental improvement of the simulation of Australian climate compared to CMIP5CMIP6 climate projections are similar to CMIP5 except for a group of higher sensitivity models with warmer projected change [ABSTRACT FROM AUTHOR]

State and local governments, businesses, community organizations, and the general public are taking an increasingly significant role in climate impact assessment. Driven by impacts to constituents and customers and threats to human health, essential services, and property values, local organizations are on the front lines of responding to climate change. National and international efforts such as the U.S. National Climate Assessment and Intergovernmental Panel on Climate Change provide fundamental scientific understanding as well as methods and modeling tools. Subnational climate assessments can build on this foundation and tailor models and analyses to specific local or decision contexts. However, subnational climate assessment and adaptation presents new scientific and research challenges, such as those related to downscaling climate models, simulating extreme events, and understanding local values and institutional practices. As state, local, and sectoral assessments become more common across the nation, there is a critical opportunity to share learnings and identify challenges and pitfalls. Building off a National Academies of Sciences, Engineering, and Medicine activity focusing exclusively on subnational climate assessment, we consider the methods and results from a variety of examples to synthesize findings about current practices and the future of subnational climate assessment. Plain Language Summary: Climate assessments provide descriptions of the current and future impacts of climate change. Such assessments have long been a part of national and international activities on climate change. Increasingly, state and local governments, businesses, and other groups have begun climate assessments that address possible impacts from climate change specific to their constituents or customers. When taking national and international efforts to a local level, two important challenges arise. One is related to the challenge of applying science and modeling developed for the large scale to a smaller scale. The second is the need to translate technical analysis into usable information for decision makers and the public. Here we build off of a workshop convened by the National Academies of Sciences, Engineering, and Medicine to consider these issues. Key Points: State and local governments and others are taking an increasingly significant role in climate impact assessmentSuch assessments present analytical challenges from downscaling national models, simulating extremes, and integrating local perspectivesThe evolution of California's assessments, in its fourth iteration, reflects the demands for climate assessments to translate science into response activities [ABSTRACT FROM AUTHOR]

Many problems in climate science require extracting forced signals from a background of internal climate variability. We demonstrate that artificial neural networks (ANNs) are a useful addition to the climate science "toolbox" for this purpose. Specifically, forced patterns are detected by an ANN trained on climate model simulations under historical and future climate scenarios. By identifying spatial patterns that serve as indicators of change in surface temperature and precipitation, the ANN can determine the approximate year from which the simulations came without first explicitly separating the forced signal from the noise of both internal climate variability and model uncertainty. Thus, the ANN indicator patterns are complex, nonlinear combinations of signal and noise and are identified from the 1960s onward in simulated and observed surface temperature maps. This approach suggests that viewing climate patterns through an artificial intelligence (AI) lens has the power to uncover new insights into climate variability and change. Plain Language Summary: Many problems in climate science require the identification of signals amidst a sea of climate "noise" and across a variety of models which can disagree with one another. Here, we demonstrate that machine learning techniques, specifically artificial neural networks, can help identify forced patterns of temperature and precipitation within climate model simulations as well as the observations. In fact, the neural network is able to identify patterns of forced change of surface temperature as early as the 1960s in climate model simulations. The results shown here are strongly suggestive of the potential power of machine learning for climate research. Key Points: Neural networks can identify forced patterns of surface temperature and precipitation amidst climate noise and model disagreementThese "indicator patterns" of forced change are present in the observationsViewing climate patterns through an AI lens has the power to uncover new insights into climate variability and change [ABSTRACT FROM AUTHOR]

As communities prepare for the impacts of climate change, policy makers and stakeholders increasingly require locally resolved projections of future climate. Statistical downscaling uses low‐resolution outputs from climate models and historical observations to both enhance the spatial resolution and correct for systematic biases. By examining the downscaled rainfall over land, we show that although bias corrections are effective in reducing biases in the current climate, they do not reduce the intermodel spread in future rainfall projections. This failure stems from the strong dependence of future rainfall change upon the current climatological rainfall patterns. Even after bias corrections are applied, the downscaled projections of precipitation change retain this dependence upon their native climatology. However, we show that this dependence can be exploited; even very simple methods to subset models according to their ability to resolve the observed rainfall climatology can substantially reduce the intermodel spread in rainfall projections. Plain Language Summary: To prepare for future climate change, policy makers and stakeholders require reliable model projections with high spatial resolution. To meet this need, statistical methods have been developed to postprocess the model output so correct for systematic biases and enhance the spatial resolution. We focus on rainfall over land and find that the postprocessing only yields consistent values for the current climate and no reduction in the uncertainty is obtained for future projections even after the postprocessing. We show that patterns of future rainfall change exhibit a strong dependence on the current climate. This dependence can be exploited by selecting models based on their ability to reproduce the observed climate. We show that this screening of models can significantly reduce the uncertainty in future rainfall projections. The success of this simple method emphasizes the importance of model evaluation in reducing the uncertainty of multimodel climate change projections. Key Points: Statistical downscaling improves the current climatology but does not reduce the intermodel spread of future rainfall projectionsModels with more similar base climates yield more similar projections of change both with and without statistical downscalingScreening models provides a simple method for constraining future projections and is insensitive to the choice of observational data sets [ABSTRACT FROM AUTHOR]

The equilibrium climate sensitivity, that is, the global‐mean surface‐air temperature change in response to a doubling of the carbon dioxide concentration is a widely used metric in climate change studies. Its exact value is rarely known because its estimation requires a long integration time of several thousand years. We propose a method to estimate an accurate value of the equilibrium response from fully coupled climate models at a reasonable computational cost. Using this method, our state‐of‐the‐art climate model CNRM‐CM6‐1 reaches a stationary state after only few hundred of years of integration. This "Fast‐Forward" method consists of an optimal two‐step forcing pathway designed using the framework of a two‐layer energy balance model. It can be applied easily to any coupled climate model. Key Points: A simple method for estimating the equilibrium climate sensitivity is proposedThe method allows to simulate the stationary climate corresponding to any given radiative perturbation with a limited computational costThe method can be applied to any atmosphere‐ocean coupled climate model [ABSTRACT FROM AUTHOR]

The Max Planck Institute Grand Ensemble (MPI‐GE) is the largest ensemble of a single comprehensive climate model currently available, with 100 members for the historical simulations (1850–2005) and four forcing scenarios. It is currently the only large ensemble available that includes scenario representative concentration pathway (RCP) 2.6 and a 1% CO2 scenario. These advantages make MPI‐GE a powerful tool. We present an overview of MPI‐GE, its components, and detail the experiments completed. We demonstrate how to separate the forced response from internal variability in a large ensemble. This separation allows the quantification of both the forced signal under climate change and the internal variability to unprecedented precision. We then demonstrate multiple ways to evaluate MPI‐GE and put observations in the context of a large ensemble, including a novel approach for comparing model internal variability with estimated observed variability. Finally, we present four novel analyses, which can only be completed using a large ensemble. First, we address whether temperature and precipitation have a pathway dependence using the forcing scenarios. Second, the forced signal of the highly noisy atmospheric circulation is computed, and different drivers are identified to be important for the North Pacific and North Atlantic regions. Third, we use the ensemble dimension to investigate the time dependency of Atlantic Meridional Overturning Circulation variability changes under global warming. Last, sea level pressure is used as an example to demonstrate how MPI‐GE can be utilized to estimate the ensemble size needed for a given scientific problem and provide insights for future ensemble projects. Key Points: The 100‐member MPI‐GE is currently the largest publicly available ensemble of a comprehensive climate modelMPI‐GE currently has the most forcing scenarios of all large ensemble projects: RCP2.6, RCP4.5, RCP8.5, and 1% CO2The power of MPI‐GE is to estimate the forced response and internal variability, including changing variability, to unprecedented precision [ABSTRACT FROM AUTHOR]

The ability of the seven CMIP5 models to simulate extreme precipitation events over Iran was evaluated using the Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks–Climate Data Record (PERSIANN‐CDR) data set. The criterion used to select the CMIP5 models was the availability of historical daily precipitation data (PERSIANN‐CDR) for the retrospective period 1983–2005, as well as future projections for the three representative concentration pathways emission scenarios (RCP2.6, RCP4.5, and RCP8.5) and spatial resolution higher than 2 × 2°. This is the first study to focus on extreme precipitation climate model simulations over Iran that includes high topography and different climates. The results show that CCSM4 has the highest correlation coefficients (CC = 0.85) and lowest root‐mean‐square error (RMSE = 73.6 mm) compared to PERSIANN‐CDR for the mean annual precipitation. However, HadGEM2‐ES shows the best (highest CCs between 0.67–0.79 and almost the lowest root‐mean‐square errors [RMSEs] compared to PERSIANN‐CDR) performance for intensity indices; MIROC5 ranked seventh (least CCs and almost the highest RMSEs) among the selected models. The results show that BCC‐CSM1‐1‐M captures maximum consecutive dry days (CDD) better than the other models. The probability matching method (PMM) is used to bias‐correct daily precipitation events from CMIP5 models with respect to the PERSIANN‐CDR estimations. All the model performances designed to capture the mean annual precipitation, as well as extreme intensity indices, improved after correction. The ensemble, constructed from the bias‐corrected model simulations using multiple linear regression (MLR), has the best performance for simulating the mean annual precipitation and extreme indices (CCs between 0.82 for consecutive wet days [CWD] and 0.93 for the mean annual precipitation) compared to the PERSIANN‐CDR estimations. Among the seven selected models, CCSM4 has the highest ranking (CCs between 0.70 for CWD to 0.91 for mean annual precipitation) after bias correction. [ABSTRACT FROM AUTHOR]

As one of three global hot spots of tropical convection, potential future changes to the Congo Basin climate system will have regional and tropics‐wide implications. However, the latest generation of climate models from the Coupled Model Intercomparison Project 5 disagree on the sign and magnitude of future change and diverge in their estimation of the historical rainfall climatology. This study assesses the plausibility of different signals of future rainfall change by examining the processes relating to rainfall projections in samples of historically wet or dry models during the September–November rainy season. In the west Congo Basin, there are no significant differences in rainfall change projections in models that are historically wet or dry. Both composites feature wetting in the north (up to 1.8 mm/day) and drying in the south, associated with enhanced tropical Atlantic sea surface temperatures, increased evaporation, and enhanced low‐level moisture flux into the basin. In the east Congo Basin, there is greater evidence that differences in model historical climatologies has an influence on the magnitude of future rainfall change. Historically wet models project significant wetting in the northeast (1.19 mm/day) associated with a weakened northern component of the African Easterly Jet (AEJ) and enhanced moisture convergence. Dry models do not capture the structure of the AEJs in the historical period, and so changes to the AEJs under warming do not produce the same wetting pattern. The analysis therefore casts doubt on the plausibility of the driest rainfall change signals in the east Congo Basin. Key Points: The magnitude and distribution of Congo Basin future rainfall change varies across coupled modelsProcess‐based model evaluation can help to identify the plausibility of rainfall projections in the absence of observational dataHistorical model biases in the African Easterly Jets affect the magnitude of future rainfall change in the east Congo Basin [ABSTRACT FROM AUTHOR]

Potential increases in the risk of extreme weather events under climate change can have significant socio‐economic impacts at regional levels. These impacts are likely to be particularly high in South Asia where Bangladesh is one of the most vulnerable countries. Regional climate models (RCMs) are valuable tools for studying weather and climate at finer spatial scales than are typically available in global climate models. Quantitative assessment of the likely changes in the risk of extreme events occurring requires very large ensemble simulations due to their rarity. The weather@home setup within the climateprediction.net distributed computing project is capable of providing the necessary very large ensembles at regionally higher resolution, but has only been evaluated over the South Asia region for its representation of seasonal climatological and monthly means. Here, we evaluate how realistically the HadAM3P‐HadRM3P model setup of weather@home can reproduce the observed patterns of temperature and rainfall in Bangladesh with focus on the modelled extreme events. Using very large ensembles of regional simulations, we find that there are substantial spatial and temporal variations in rainfall and temperature biases compared with observations. These are highest in the pre‐monsoon, which are largely caused by timing issues in the model. Modelled mean monsoon and post‐monsoon temperatures are in good agreement with observations, whereas there is a dry bias in the modelled mean monsoon rainfall. The rainfall bias varies both spatially and with the data set used for comparison. Despite of these biases, the model‐simulated temperature and rainfall extremes in summer monsoon over Bangladesh are approximately representative of the observed ones. At the wettest parts of northeast Bangladesh, rainfall extremes are underestimated compared to GPCC and APHRODITE but are within the range of CPC observations. Therefore, the weather@home RCM, HadRM3P may provide a sufficiently reliable tool for studying the extreme weather events in Bangladesh. [ABSTRACT FROM AUTHOR]

Coupled Model Intercomparison Project phase 5 (CMIP5) climate models simulate a wide range of historical sea ice areas. Even models with areas close to observed values may contain compensating errors, affecting reliability of their projections. This study focuses on the seasonal cycle of sea ice, including analysis of model concentration budgets. Many models have insufficient autumn ice growth, leading to large winter biases. A subset of models accurately represent sea ice evolution year‐round. However, comparing their winter ice concentration budget to observations reveals a range of behaviors. At least one model has an accurate ice budget, which is only possible due to realistic ice drifts. The CMIP5 generation of model physics and resolution is therefore structurally capable of accurately representing processes in Antarctic sea ice. This implies that substantially improved projections of Antarctic dense ocean water formation and ice sheet melting are possible with appropriate subsetting of existing climate models. Plain Language Summary: Antarctic sea ice simulated by numerical climate models generally exhibits large differences compared to satellite‐based estimates. Therefore, attempts to use these models to understand how Antarctic sea ice—and as a result Antarctic climate, sea level rise, and ocean circulation—will change in the future are severely compromised. This study looks beyond standard measures of sea ice such as the area of ocean covered by ice. Instead, we calculate how much of the change in sea ice over the course of a year is caused by local melting and freezing on the one hand and ice transport (for example by winds and ocean currents) on the other. Of an already small number of models which do show a realistic amount of sea ice, we identify that some do so for the wrong reasons. However, there is one model using established model components which shows realistic behavior. This model is able to match observations of sea ice drift. This demonstrates that the existing climate models are capable of representing Antarctic sea ice without further increases in sophistication. Focusing on such models for climate projections may improve the reliability of projections of the future. Key Points: Some CMIP5 models capture Antarctic sea ice area, but quantifying their dynamic and thermodynamic processes reveals serious deficienciesOne model is right for the right reasons; realistic Antarctic sea ice can be simulated using existing model physics and resolutionRealistic representation of sea ice processes in this model is only possible because it has realistic sea ice drift [ABSTRACT FROM AUTHOR]

We advance a simple framework based on "time of detection" for estimating the observational needs of studies assessing climate changes amidst natural variability and apply it to several examples related to ocean acidification. This approach aims to connect the Global Ocean Acidification Observing Network "weather" and "climate" data quality thresholds with a single dynamic threshold appropriate for a range of potential ocean signals and environments. A key implication of the framework is that measurement frequency can be as important as measurement accuracy, particularly in highly variable environments. Pragmatic cost‐benefit analyses based on this framework can be performed to quantitatively determine which observing strategy will accomplish a given detection goal soonest and resolve a signal with the greatest confidence and to assess how the trade‐offs between measurement frequency and accuracy vary regionally. Key Points: Time of detection can be used to evaluate the effectiveness of climate observing effortsMeasurement frequency can be as important as measurement accuracy, particularly in highly variable environmentsClimate quality measurements are critical for monitoring some, but not all, climate signals [ABSTRACT FROM AUTHOR]

A nonlinear method has been developed to estimate climate feedbacks based on the Neural Network (NN) taking advantage of its self‐learning skills. The NN model developed here is trained using a reanalysis data set and predicts radiation flux globally from atmospheric and surface variables. The radiative feedbacks of temperature, water vapor, surface albedo, and cloud in the interannual climate variations estimated from the NN method are in agreement with those from a broadly used kernel method. However, the NN method demonstrates significant advantages: (1) it withdraws the linearity assumption of the kernel method and accounts for the nonlinear effects of the feedbacks. In the case of large climate perturbations, such as that in the Arctic caused by sea ice melt, the NN method achieves better radiation closure. (2) The method can directly calculate the radiative feedback of cloud and its components. We find that the high, middle, and low cloud feedback components analyzed from the NN method are linearly additive in the interannual climate variations, although there is a considerable nonlinear effect arising from the interactions between cloud and noncloud variables. Key Points: A nonlinear Neural Network (NN) method has been developed to estimate climate feedbacksThe NN method accounts for the nonlinear effects of the feedbacks and achieves good radiation closureThe NN method can directly calculate the radiative feedback of cloud and its components [ABSTRACT FROM AUTHOR]

A Regional Extreme Climatic Change Index (RECCI), simultaneously considering the changes in intensity, frequency and interannual variability of three major extreme climatic variables (i.e., precipitation, temperature and wind speed), is constructed to represent regional changes of climate extremes in response to global warming. First, the daily outputs from 13 models in the Coupled Model Intercomparison Project phase 5 project in both historical and future simulations under the Representative Concentration Pathway 8.5 scenario are used to compute the extreme climatic indices. Second, the RECCI is computed on both annual and seasonal time scales during three periods (i.e., 2016–2035, 2046–2065 and 2080–2099) over 26 subregions. Finally, the spatiotemporal change of the RECCI is investigated, and then, the 26 subregions are classified into four categories for each period. The first category with the largest RECCI value is very sensitive to global warming, which is called hot spots of climate extremes. The results show that most hot spots are not time invariant on annual and seasonal time scales with some exceptions. On the annual time scale, the Amazon Basin is the only persistent hot spot in all three periods. For the seasonal time scale in March‐April‐May, the climate extremes in the Amazon Basin always display the strongest responsiveness to global warming, and the Eastern Africa is the only persistent hot spot in June‐July‐August in three periods. Similar results are also found for the other two seasons and periods. In addition, the change in extreme temperature is crucial over the East Asia with change in frequency prominent. Key Points: The majority of the regions in different categories of climate extremes are not time invariant on annual and seasonal time scalesThe Amazon Basin is the only persistent hot spot of climate extremes in 2016–2035, 2046–2065 and 2080–2099The Regional Extreme Climatic Change Index is a robust metric that is not sensitive to the actual models chosen [ABSTRACT FROM AUTHOR]

Retrospective tropical Indo‐Pacific forecasts for 1961–2015 are made using 28 models from the fifth phase of the Coupled Model Intercomparison Project (CMIP5) plus four models from the North American Multi‐Model Ensemble (NMME), using a model‐analog technique. Forecast ensembles are extracted from preexisting model simulations, by finding those states that initially best match an observed anomaly and tracking their subsequent evolution, requiring no additional model integrations. Model‐analog forecasts from the 10 "best" CMIP5 models have skill for sea surface temperature and precipitation comparable to that of both the NMME model‐analog forecast ensemble and (since 1982) traditional assimilation‐initialized NMME hindcasts. The El Niño–Southern Oscillation (ENSO) forecast skill has no trend over the 55‐year period, and its decadal variations appear largely random, although the skill does improve during epochs of increased ENSO activity. Including the CMIP5‐projected effects of external radiative forcings improves the tropical sea surface temperature skill of the model‐analog forecasts but not within the ENSO region. Plain Language Summary: Seasonal forecasts are made by starting a climate model from an initial estimate of the latest global three‐dimensional ocean, atmosphere, and land conditions and then using supercomputers to run the model's equations forward in time. These extensive calculations are only feasible at a few national operational centers and large research institutions. However, many similar models are also used for long simulations of the Earth's preindustrial climate, which are made freely available for climate change studies. We investigated whether seasonal forecasts might be drawn from the information already existing within these simulations, instead of by making new model computations. Within each simulation, we determine the best matches, or "model‐analogs," to current observed tropical Indo‐Pacific ocean surface conditions. How these analogs evolve over the next several months within each long simulation is then its seasonal forecast. We find that this much less expensive model‐analog technique is as skillful as the more traditional forecasting method. We then employed it to make forecasts during 1961–2015, using 28 existing climate simulations, significantly expanding previous forecast efforts. This study suggests that with little additional effort, sufficiently realistic and long existing climate model simulations can provide the basis for skillful seasonal forecasts. That is, anyone can be a climate forecaster. Key Points: Seasonal tropical Indo‐Pacific hindcasts for 1961‐2015 are made from 28 different CMIP5 models, with some as skillful as operational CGCMsENSO hindcast skill has no trend, but it is higher in strong‐ENSO epochs (1980s and 1990s) and lower in weak‐ENSO epochs (1970s and 2000s)Including the effects of projected external radiative forcings improves SST hindcast skill but not within the ENSO region [ABSTRACT FROM AUTHOR]

This study explores the extent to which convective aggregation interacts with sea surface temperature (SST) and affects climate sensitivity. For this purpose, radiative‐convective equilibrium simulations are run with a general circulation model coupled to an ocean mixed layer, and several types of perturbations are imposed to the ocean‐atmosphere system. Convective aggregation turns out to be much more sensitive to temperature in coupled experiments than in prescribed SST experiments. But changes in convective aggregation induced by a doubling of the CO2 concentration are always smaller than changes associated with the transition from a non‐aggregated to an aggregated state. If aggregation changes were acting alone, they would exert a strong negative feedback on global mean surface temperature. However, in a coupled framework, aggregation changes interact with the SST and generate SST gradients that strengthen the positive low‐cloud feedback associated with changes in SST pattern. This overcompensates the negative feedback due to aggregation changes and leads to a larger equilibrium climate sensitivity than in the absence of SST gradients. Although this effect might be model specific, interactions between convective aggregation and the spatial distribution of SST appear crucial to assess the impact of convective aggregation on climate sensitivity. Key Points: Convective aggregation interacts strongly with surface temperature gradientsAggregation can increase climate sensitivity when coupled with an interactive slab ocean, rather than decreasing itThe impact of convective aggregation on climate should also be investigated in coupled frameworks [ABSTRACT FROM AUTHOR]

Abstract: Glacial‐interglacial cycles are often described as an amplified global response of the climate to perturbations in solar radiation caused by oscillations of Earth's orbit. However, it remains unclear whether internal feedbacks are large enough to account for the radically different glacial and interglacial states. Here we provide support for an alternative view: Glacial‐interglacial states are multiple equilibria of the climate system that exist for the same external forcing. We show that such multiple equilibria resembling glacial and interglacial states can be found in a complex coupled general circulation model of the ocean‐atmosphere‐sea ice system. The multiple states are sustained by ice‐albedo feedback modified by ocean heat transport and are not caused by the bistability of the ocean's overturning circulation. In addition, expansion/contraction of the Southern Hemisphere ice pack over regions of upwelling, regulating outgassing of CO2 to the atmosphere, is the primary mechanism behind a large pCO2 change between states. [ABSTRACT FROM AUTHOR]

Abstract: On interannual to decadal time scales, the climate mode with many of the strongest societal impacts is the El Niño–Southern Oscillation (ENSO). However, quantifying ENSO's changes in a warming climate remains a formidable challenge, due to both the noise arising from internal variability and the complexity of air‐sea feedbacks in the tropical Pacific Ocean. In this work, we use large (≥30‐member) ensembles of climate simulations to show that anthropogenic climate change can produce systematic increases in ENSO teleconnection strength over many land regions, driving increased interannual variability in regional temperature extremes and wildfire frequency. As the spatial character of this intensification exhibits strong land‐ocean contrasts, a causal role for land‐atmosphere feedbacks is suggested. The identified increase in variance occurs in multiple model ensembles, independent of changes in sea surface temperature variance. This suggests that in addition to changes in the overall likelihoods of heat and wildfire extremes, the variability in these events may also be a robust feature of future climate. [ABSTRACT FROM AUTHOR]

Abstract: Previous work has explored the linkages between Arctic sea ice extent (SIE) anomalies at the end of the summer melt season and high‐latitude climate. Here we show that Arctic midsummer SIE anomalies provide predictive skill on time scales of ~2–3 months for high‐latitude climate. Midsummers characterized by low SIE are associated with significant positive temperature and easterly wind anomalies throughout the high‐latitude troposphere through September and significant positive temperature anomalies at the Arctic surface into October. The inferred predictive skill for autumn climate derives from the persistence of the sea ice field. It is robust throughout the Arctic basin and is supported in climate models from the fifth phase of the Coupled Model Intercomparison Project archive and in prediction experiments from the Arctic Predictability and Prediction on Seasonal to Interannual Time scales project. It is theorized that the predictive skill derives from (1) the anomalous storage of heat in the Arctic Ocean during periods of low summertime SIE and (2) the delayed formation of sea ice during the following autumn months. [ABSTRACT FROM AUTHOR]

Climate inter‐annual variability over the North American monsoon (NAM) region is associated with El Niño‐Southern Oscillation (ENSO) and Pacific decadal variability (PDV), which drive a warm season atmospheric teleconnection response. Using the North American Regional Climate Change Assessment Program (NARCCAP) simulations, previous studies have found that regional models forced with an atmospheric reanalysis (NARCCAP Phase I) represent the NAM reasonably well as a climatological feature. However, when these same regional models are forced with global climate model projections (NARCCAP Phase II), their ability to represent the NAM as a salient feature substantially degrades. The present study evaluates NAM inter‐annual climate variability through the continental‐scale patterns of summer precipitation within the NARCCAP simulations (Phases I and II), in relation to ENSO–PDV, and the presence of the driving atmospheric teleconnection response. Multivariate statistical analyses are applied to sea surface temperature and precipitation data sets to determine dominant variability at continental scale, with focus on the southwest. The analysis reveals that NARCCAP Phase I simulations are able to portray the spatial pattern of precipitation associated with ENSO–PDV in a similar way to observations. However, all NARCCAP Phase II simulations, with the exception of the HRM(Hadcm3) regional–global model pair, fail to reproduce this climate variability. Although including all possible NARCCAP model simulations to generate a multi‐model ensemble mean would increase the statistical degree of confidence in climate projections, this type of result would not increase confidence in the physical climatology of model representations of warm season climate variability. More physically based, process‐oriented metrics are needed to evaluate model quality in assessing the uncertainty of future climate change in multi‐model ensemble products used for climate change impacts assessments. [ABSTRACT FROM AUTHOR]

This study demonstrates a methodology to construct short‐duration rainfall intensity–duration–frequency (IDF) curves and to quantify the variability in the rainfall intensities for different return periods with respect to the changing climate. A dynamical downscaling approach using the regional climate model (RCM) Weather Research and Forecasting (WRF) has been used to assess present and future climates using the downscaling of an ensemble of three global climate models (GCMs) (CSIRO‐ACCESS1.3, MPI‐ESM‐MR ECHAM6 and NIES‐MIROC5) under the Coupled Model Intercomparison Project phase 5 (CMIP5). Furthermore, a statistical approach using the well‐known simple scaling method has been applied to extend the 6‐hourly WRF precipitation output to the finer temporal scale of 10 min. The short‐duration IDF curves were then constructed for the present and future climates under two representative concentration pathway (RCP) scenarios RCP4.5 and RCP8.5. A preliminary examination for this case study over Bac Ninh, an industrial area in the northern Vietnam, shows that there is a substantial increase in short‐duration rainfall intensity in the future with respect to the baseline climate. The highest increase is towards the end of the century (2071–2100) ranging from 56 to 61% for a 10‐ and 100‐year return period for 24‐hr duration, respectively, while the increase is about 40–45% for the 10‐min duration. These results strongly suggest that severe flooding in the future climate over the study region may be likely. The study results might be useful for policymakers and infrastructure planning and for insurance companies around the study area. [ABSTRACT FROM AUTHOR]

Simulations of the regional climate model REMO at 25 and 50 km (0.44° and 0.22°, respectively) grid spacing are evaluated over Central Africa for the period 1998-2010. The model's ability to simulate various aspects of the observed climate is assessed with a main focus on precipitation characteristics. The results show that the REMO model has a good skill to reproduce the main regional features of the mean climate fields as well as seasonal and annual cycles, although some biases still exist. The model captures the variability in precipitation anomalies between different events associated with El Niño/Southern Oscillation. Generally, the findings remain largely insensitive to the refinement of the horizontal resolution, especially when considering the regional scale. This points the fact that the model dynamical and physical setup may have to be adjusted when setting a focus on the Central African region, in order to better represent the climate mechanisms. It is further found that REMO is capable of capturing the basic features of climatic extremes and accurately represents the shape of the frequency distribution of daily precipitation data. Daily characteristics of precipitation are simulated much better by REMO at 25 km. [ABSTRACT FROM AUTHOR]

ABSTRACT Under the framework of an project 'Building Asian Climate Change Scenarios by Multi-Regional Climate Models Ensemble', the ability of eight regional climate models and two fine-resolution global climate models to reproduce late 20th century (1981-2000) precipitation climatology is assessed. Future precipitation change (2041-2060) under the A1B scenario is also quantified by applying four different ensemble methods: equal weighting, weighted mean ( WM), reliability ensemble averaging ( REA) and performance-based ensemble averaging, after applying fourfold cross-validation using observation and multi-model-simulated precipitation. The results indicate that the ensemble of simulated precipitation outperforms any single RCM in many aspects. Among the four ensemble approaches, the WM and REA methods show better skill in improving the simulation results, and are used for ensemble prediction of regional climate in Asia. Under the A1B scenario, the WM method estimates future precipitation change of approximately 0.2 mm day−1 with less precipitation in northern and western China and northern India, and more precipitation in most other areas in Asia. The future annual precipitation will decrease by 0.1-0.5 mm day−1 in northern India, Pakistan and the central area of southern China. No significant change is found over eastern Kazakhstan, Mongolia, north-central and western China. [ABSTRACT FROM AUTHOR]

ABSTRACT An ensemble of simulations from four regional climate models ( RCMs) driven by a global reanalysis was obtained from the North American Regional Climate Change Assessment Program ( NARCCAP) and used to evaluate the ability of the RCMs to simulate the long-term (1979-2000) climatology of southerly low-level jets (S- LLJs) in the central United States. The RCM-derived S- LLJ climatologies were evaluated against rawinsonde observations for the same period. The use of a small ensemble of RCM simulations helped to identify model differences and assisted with interpretation. The RCMs generally reproduced the broad spatial patterns and temporal variations of jet frequency and average jet height and speed. No model consistently outperformed the others in all aspects of the evaluation, although differences existed between models in the placement, migration and relative strength of 'hotspots' of more frequent jet activity. In particular, three of the four models placed the centre of greatest nocturnal S- LLJ activity during the warm season in northern and central Texas, whereas for the other model the greatest jet activity was located in the south-central plains (Kansas/Oklahoma). The magnitude of a S- LLJ frequency maximum over south Texas also varied between models, with simulated frequencies exceeding observed frequencies for some models but substantially underestimating for others. The evaluation presented here highlights the potential applications of RCMs in S- LLJ research for future climate and other assessment studies that require three-dimensional data with relatively high spatial and temporal resolutions. The overall performance of the models in reproducing the long-term S- LLJ climatology supports the use of NARCCAP RCM simulations in climate assessments for the central United States where S- LLJs are an important contributor to the regional climatology. [ABSTRACT FROM AUTHOR]

This is the first of three parts of an introduction to the philosophy of climate science. In this first part about observing climate change, the topics of definitions of climate and climate change, data sets and data models, detection of climate change, and attribution of climate change will be discussed. [ABSTRACT FROM AUTHOR]

This is the second of three parts of an introduction to the philosophy of climate science. In this second part about modelling climate change, the topics of climate modelling, confirmation of climate models, the limits of climate projections, uncertainty and finally model ensembles will be discussed. [ABSTRACT FROM AUTHOR]

ABSTRACT Extreme climate events over China at the end of the 21st century (2080-2099) are investigated using the regional climate model RegCM4. Model performance is validated through comparison between observations and simulations over the period 1985-2005. The results show that RegCM4 can satisfactorily reproduce the spatial distribution of extreme climate events over China. The model simulates temperature extremes more accurately than precipitation. Under the RCP8.5 (high emission) scenario, the number of frost days decreases, and both the heat wave duration index and the growing season length increase dramatically towards the end of the 21st century. Changes in extreme temperature become increasingly pronounced from South to North China, with the most significant changes occurring on the Tibetan Plateau ( TP). The proportion of heavy precipitation generally increases, except on the southern TP. The number of very heavy precipitation days increases by 25-50% in Northwest and East China. In winter, the number of consecutive dry days ( CDD) decreases in North China and increases in South China. The greatest increases in CDD are found in June, July and August ( JJA) in Southwest China. In a future that follows this scenario, drought events may be aggravated in Southwest China, and decrease in North China. In contrast, when repeating these projections under the assumption of the RCP4.5 scenario for emissions, the frequency of extreme climate events is reduced. These results suggest that reductions in greenhouse gas emissions may mitigate the effects of climate change over the coming decades. [ABSTRACT FROM AUTHOR]

This study applies the factor separation ( FS) method to investigate the contributions of different factors, along with their synergy, on a set of regional climate model ( RCM) projections for the Mediterranean region. The FS method is applied to six projections for the period 1970-2100 performed with the regional model RegCM4 over the Med- CORDEX domain. Two different sets of factors are intercompared, namely the driving global climate model ( GCM) boundary conditions against two model physics settings (convection scheme and irrigation). We demonstrate the usefulness of the FS method to assess different sources of uncertainty in RCM-based regional climate projections. [ABSTRACT FROM AUTHOR]