Search

Climate trends have been observed over the recent decades in many parts of the world, but current global climate models (GCMs) for seasonal climate forecasting often fail to capture these trends. As a result, model forecasts may be biased above or below the trendline. In our previous research, we developed a trend-aware forecast postprocessing method to overcome this problem. The method was demonstrated to be effective for embedding observed trends into seasonal temperature forecasts. In this study, we further develop the method for postprocessing GCM seasonal precipitation forecasts. We introduce new formulation and evaluation features to cater for special characteristics of precipitation amounts, such as having a zero lower bound and highly positive skewness. We apply the improved method to calibrate ECMWF SEAS5 forecasts of seasonal precipitation for Australia. Our evaluation shows that the calibrated forecasts reproduce observed trends over the hindcast period of 36 years. In some regions where observed trends are statistically significant, forecast skill is greatly improved by embedding trends into the forecasts. In most regions, the calibrated forecasts outperform the raw forecasts in terms of bias, skill, and reliability. Wider applications of the new trend-aware postprocessing method are expected to boost user confidence in seasonal precipitation forecasts.

Comparisons of precipitation between general circulation models (GCMs) and observations are often confounded by a mismatch between model output and instrument measurements, including variable type and temporal and spatial resolutions. To mitigate these differences, the radar-simulator Quickbeam within the Cloud Feedback Model Intercomparison Project (CFMIP) Observation Simulator Package (COSP) simulates reflectivity from model variables at the subgrid scale. This work adapts Quickbeam to the dual-frequency precipitation radar (DPR) on board the Global Precipitation Measurement (GPM) satellite. The longer wavelength of the DPR is used to evaluate moderate to heavy precipitation in GCMs, which is missed when Quickbeam is used as a cloud radar simulator. Latitudinal and land–ocean comparisons are made between COSP output from the Community Atmospheric Model version 5 (CAM5) and DPR data. Additionally, this work improves the COSP subgrid algorithm by applying a more realistic, nondeterministic approach to assigning GCM gridbox convective cloud cover when convective cloud is not provided as a model output. Instead of assuming a static 5% convective cloud coverage, DPR convective precipitation coverage is used as a proxy for convective cloud coverage. For example, DPR observations show that convective rain typically only covers about 1% of a 2° grid box, but that the median convective rain area increases to over 10% in heavy rain cases. In our CAM5 tests, the updated subgrid algorithm improved the comparison between reflectivity distributions when the convective cloud cover is provided versus the default 5% convective cloud-cover assumption.

The accuracy of the climate impact models depends on the representative power of the climate model used in the analysis. Many problems in these impact models can be eliminated by the selection of the appropriate climate models as an input. The main purpose of this research is to find a methodology to select the more representative climate models. In recent years, the quantity and quality of climate models have significantly improved. In several projects, many different Global Climate Model (GCM)–Regional Climate Model (GCM) combinations have been carried out. One of these recent comprehensive projects is CORDEX. The monthly temperature outputs of various GCM-RCM combinations over Europe between the years 1976 and 2005 were used in this paper for the determination of the more representative combinations. E-OBS is used as the observation data set in this research. Using the proposed method, a clear difference between the more, moderately, and less representative model combinations could be obtained. To model the future temperature climatology of Europe, the best choice of the analysis method is using MPI as GCM and either REMO or CCLM5 as RCM over the whole domain. According to our analysis, the best representative climate model combination can be performed by running MPI-CCLM5 combination over the EURO-CORDEX domain.

Observational studies show a strong connection between the intraseasonal Madden-Julian oscillation (MJO) and the stratospheric quasi-biennial oscillation (QBO): the boreal winter MJO is stronger, more predictable, and has different teleconnections when the QBO in the lower stratosphere is easterly versus westerly. Despite the strength of the observed connection, global climate models do not produce an MJO-QBO link. Here the authors use a current-generation ocean-atmosphere coupled NASA Goddard Institute for Space Studies global climate model (Model E2.1) to examine the MJO-QBO link. To represent the QBO with minimal bias, the model zonal mean stratospheric zonal and meridional winds are relaxed to reanalysis fields from 1980-2017. The model troposphere, including the MJO, is allowed to freely evolve. The model with stratospheric nudging captures QBO signals well, including QBO temperature anomalies. However, an ensemble of nudged simulations still lacks an MJO-QBO connection.

Marine Heatwave (MHW) events have been increasing all around the world, causing severe impacts on marine ecosystems and on the economy of the aquaculture, fishing and tourism industries. In this study, the occurrence and characteristics of MHW events in the North Atlantic are analyzed for a recent period (1971–2000) and the two future periods (2041–2070 and 2071–2100). The analysis is based on Sea Surface Temperature (SST) data obtained from observations and from Global Climate Models (GCM) in the CMIP5 archieve. The results show that the GCMs present significant shortcomings in reproducing the SST and MHWs. The model results regarding the recent trends in MHW are in agreement with the ones observed, although they underestimate the values of MHW intensity and overestimate the values of both frequency and duration of events. The MHWs observed occur at a mean frequency of 1.90 events per year and are characterized by ~ 13 days of mean duration and 0.37 oC of mean intensity, while the multi-model ensemble mean characterizes the events with 12 additional days and 0.15 oC less intensity. Under climate change scenarios, when considering a stationary threshold, the models project noticeable increases in MHW event intensity that could reach 2 oC above the 90th percentile and a quasi-permanent state of MHW by the end of the century. When a non-stationary threshold is considered, the characteristics of the events are similar to those obtained during the historical period. The GCMs significant biases in simulating the SST and leading to extremes like MHWs highlight the importance of improving GCM performance.

Current rainfall Intensity-Duration-Frequency (IDFs) curves generally were developed based on stationary extreme value assumption. However, climate change is expected to alter extreme rainfalls and invalidate the stationary assumption. So, it is crucial to develop future rainfall IDFs taking into account the impacts of climate change. Difficulty in preparing reliable rainfall time series with fine time step and capturing extremes for future climate scenarios has led to limited studies aimed at investigating the change of IDF curves due to climate change. In this paper, a robust stochastic rainfall model (NSRP) is employed to produce future rainfall IDFs for five stations across Karkheh and Karun basins in Iran. NSRP can generate long-term realistic rainfall series containing extremes. Moreover, it applies changes in different rainfall statistical characteristics, projected by GCMs, in the downscaling procedure. For each station, the model was calibrated using observed rainfalls series. Then, 3000-year daily rainfall series were generated, and historical IDFs were developed. Consequently, NSRP parameters were perturbed based on GCM projections. Then, 3000-year future rainfall series were generated, and future IDFs were developed. The climate change uncertainties were represented by employing two GCM (CanESM2 and HadGem2) and three emission scenarios (RCP2.6, RCP4.5, and RCP8.5). The results showed NSRP is able to reproduce observed extreme rainfalls in a wide range of time scales. Also, climate change will lead to a considerable increase in future extreme rainfall intensity in the study basins. As averages of all considered scenarios, rainfall intensities will increase between 22 and 206% in the future.

This study compares radiation-circulation coupling over tropical and subtropical oceans by examining ice water path, radiation, low-level wind, and precipitation fields from the uncoupled prescribed sea surface temperature AMIP6 and the fullycoupled CMIP6 historical runs. Ice water path of the CMIP6 ensemble shows a closer agreement than the AMP6 ensemble, particularly in the subtropics. The inclusion of falling ice (snow) radiative effects (FIREs), in general, improves simulation of radiation, low-level wind, and precipitation fields over the northwest Pacific, Southeast Pacific Convergence Zone (SPCZ), equatorial eastern Pacific and Atlantic Intertropical Convergence Zone (ITCZ). When both AMIP6 and CMIP6 models are divided into two groups with inclusion and exclusion of FIREs, the impact of FIREs is most pronounced in ITCZ and the subtropical trade-wind regions in CMIP6 but over the tropical Pacific and SPCZ in AMIP6. This suggests that active ocean plays a significant role in radiation-circulation coupling. The CMIP6 models with FIREs have less over-estimated biases in upward longwave radiation over the convective zones in Pacific and Atlantic and less low-level divergence of anomalous flows over convective zones, i.e., stronger trade winds. The circulation changes stronger in CMIP6 over the trade wind regions than in AMIP6, which suggests that the role of active ocean is important in testing an improved physical process in models. This conclusion is also supported by more systematic improvements in groups of models with inclusion of FIREs in CMIP6 than in AMIP6, hinting the important roles of FIREs in radiationcirculation coupling.

The falling ice (snow) radiative effects (FIREs) have previously been shown to contribute substantially to reduced discrepancies in simulations of present-day climatology of radiation, skin temperatures and sea ice concentration and thickness over the Southern Ocean. This study extends to examine the impacts of FIREs on simulation of sea ice changes under a scenario of gradual increase of atmospheric CO2 concentration. We perform a pair of sensitivity experiments including (CESM1-SoN) and excluding (CESM1-NoS) FIREs using Community Earth System Model version 1. The differences in the annual and seasonal means between the initial and warmer periods are examined. Relative to CESM1-SoN, CESM1-NoS simulates more surface reflected shortwave and less downward longwave radiative warming, as well as colder surface temperature, resulting in larger annual-mean sea ice extent and thickness and slower seasonal and long-term sea ice melting and thinning. Over the Southern Ocean of CESM1-SoN, reduced downwelling longwave radiation in austral winter (June-July-August: JJA) leads to sea-ice growth with colder skin temperature while reduced net radiation resulting from increased shortwave reflection in austral summer reduces the melting of sea ice with little change in skin temperature. CESM1-NoS shows seasonal and long-term trends similar to those in CMIP5 models that exclude FIREs, hinting slower future warming-driven changes and larger amplitude of the annual cycle in sea ice concentration and thickness. The ice-free Southern Ocean in peak melting season is simulated at approximately year 130 for CESM1-NoS but year 100 for CESM1-SoN, about 30 years later than that of the Arctic.

Statistical prediction of tropical sea surface temperatures (SSTs) is performed using linear inverse models (LIMs) that are constructed from both observations and general circulation model (GCM) output of SST. The goals are to establish a baseline for tropical SST predictions, to examine the extent to which the skill of a GCM-derived LIM is indicative of that GCM’s skill in forecast mode, and to examine the linkages between mean state bias and prediction skill. The observation-derived LIM is more skillful than a simple persistence forecasts in most regions. Its skill also compares well with some GCM forecasts except in the equatorial Pacific, where the GCMs are superior. The observation-derived LIM is matched or even outperformed by the GCM-derived LIMs, which may be related to the longer data record available for GCMs. The GCM-derived LIMs provide a fairly good measure for the skill achieved by their parent GCMs in forecast mode. In some cases, the skill of the LIM is actually superior to that of its parent GCM, indicating that the GCM predictions may suffer from initialization problems. A weak-to-moderate relation exists between model mean state error and prediction skill in some regions. An example is the eastern equatorial Atlantic, where an erroneously deep thermocline reduces SST variability, which in turn affects prediction skill. Another example is the equatorial Pacific, where skill appears to be linked to cold SST biases in the western tropical Pacific, which may reduce the strength of air–sea coupling.

The climate features over Central America and the Caribbean are simulated with the regional climate model RegCM4 to analyze the performance of the model in reproducing precipitation and temperature patterns over the region. Results from RegCM4 and the driving global climate model (GCM) HadGEM2-ES (HadG) are compared against gridded observations for a period from January 1980 to December 2004 using a perturbed physics ensemble formed by four configurations of the RegCM4 using different combinations of four convection schemes run over land and ocean areas: Kain Fritsch (Kf), Emanuel (Em), Grell (Ge), and Tiedtke (Tk). The added value of RegCM4 relative to the GCM simulation is also estimated, with focus over four subregions, using several metrics calculated on the RegCM4 as well as the GCM grids. The RegCM4 downscaling, expressed by the ensemble mean ENS, adds significant details to the HadG simulation of the temperature field patterns, showing smaller biases of up to ± 1.6 °C at the RCM resolution. Regarding the spatial patterns of precipitation, the HadG has overall higher correlation values with observations than ENS. However, the regional model provides more detailed spatial distributions, decreasing the bias by more than 1 mm/day over some of the islands. It also captures extreme precipitation events that are underestimated in the HadG simulation, even after upscaling to the GCM resolution grid. Concerning the performance of the different RegCM4 convection scheme combinations, KfEm and EmEm show the highest skill values for precipitation, while for temperature TkEm and GeEm are the best performing. This highlights that no individual scheme outperforms the others in all respects, while the application of the averaged ensemble technique provides the best results.

Comparing outputs of multiple climate and crop models is an option to assess the uncertainty in simulations in a changing climate. The use of multiple wheat models under five plausible future simulated climatic conditions is rarely found in literature. CERES-Wheat, DSSAT-Nwheat, CROPSIM-Wheat, and APSIM-Wheat models were calibrated with observed data form eleven sowing dates (15 October to 15 March) of irrigated wheat trails at Faisalabad, Pakistan, to explore close to real climate changing impacts and adaptations. Twenty-nine GCM of CMIP5 were used to generate future climate scenarios during 2040–2069 under RCP 8.5. These scenarios were categorized among five climatic conditions (Cool/Wet, Cool/Dry, Hot/Wet, Hot/Dry, Middle) on the basis of monthly changes in temperature and rainfall of wheat season using a stretched distribution approach (STA). The five GCM at Faisalabad and Layyah were selected and used in the wheat multimodels set to CO2 571 ppm. In the future, the temperature of both locations will elevate 2–3 °C under the five climatic conditions, although Faisalabad will be drier and Layyah will be wetter as compared with baseline conditions. Climate change impacts were quantified on wheat sown on different dates, including 1 November, 15 November, and 30 November which showed average reduction at semiarid and arid environment by 23.5%, 19.8%, and 31%, respectively. Agronomic and breeding options offset the climate change impacts and also increased simulated yield about 20% in all climatic conditions. The number of GCMs was considerably different in each quadrate of STA, showing the uncertainty in possible future climatic conditions of both locations. Uncertainty among wheat models was higher at Layyah as compared with Faisalabad. Under Hot/Dry and Hot/Wet climatic conditions, wheat models were the most uncertain to simulate impacts and adaptations. DSSAT-Nwheat and APSIM-Wheat were the most and least sensitive to changing temperature among years and climatic conditions, respectively.

The regional climate of the Arabian Gulf region is modeled using a set of simulations based on the Weather Research and Forecasting (WRF) Model, including a 30-yr benchmark simulation driven by reanalysis data, and two bias-corrected Community Earth System Model (CESM)-driven (BCD) WRF simulations for retrospective and future periods that both include 10-yr convection-permitting nested simulations. The modeled precipitation is cross-validated using Tropical Rainfall Measuring Mission data, rain gauge data, and the baseline dataset from the benchmark simulation. The changes in near-surface temperature, precipitation, and ambient conditions are investigated using the BCD WRF simulations. The results show that the BCD WRF simulation well captures the precipitation distribution, the precipitation variability, and the thermodynamic properties. In a warmer climate under the RCP8.5 scenario around the year 2070, the near-surface temperature warms by ~3°C. Precipitation increases over the Arabian Gulf, and decreases over most of the continental area, particularly over the Zagros Mountains. The wet index decreases while the maximum dry spell increases in most areas of the model domain. The future changes in precipitation are determined by both the thermodynamics and dynamics. The thermodynamic impact, which is controlled by the warming and moistening, results in more precipitation over the ocean but not over the land. The dynamic impact, which is controlled by changes in the large-scale circulation, results in decrease in precipitation over mountains. The simulations presented in this study provide a unique dataset to study the regional climate in the Arabian Gulf region for both retrospective and future climates.

Global climate models (GCMs) are used by major climate centers worldwide for global climate forecasting, and predictive performance is one of the most important issues in GCM forecast applications. In addition to spatial plotting that illustrates anomaly correlation at individual grid cells, this study proposes a novel local indicator of spatial association (LISA) of anomaly correlation (herein, LISAAC) for GCM seasonal forecasts of global precipitation. LISAAC is built upon local Moran’s I by relating anomaly correlation at neighboring grid cells to one another. While local Moran’s I takes the grand mean of anomaly correlation as the benchmark, LISSAC considers the original value of anomaly correlation in the mathematical formulation. A case study is devised for the Climate Forecast System version 2 (CFSv2) seasonal forecasts, which are initialized in January, February,…, and June, of the global precipitation in June, July, and August. Three metrics—LISAAC, local Moran’s I, and original anomaly correlation—are applied to investigate the predictive performance. In comparison with local Moran’s I, LISAAC can identify clusters of positive, neutral, and negative anomaly correlations. In comparison with anomaly correlation, LISAAC can capture outliers of positive (negative) anomaly correlation surrounded by negative (positive) anomaly correlation. Overall, the results highlight that LISAAC can serve as a useful tool for evaluating the predictive performance of GCM seasonal forecasts of global precipitation.

The precipitation behaviour at different timescales (interannual, intraseasonal and synoptic) and extreme events in a warmer scenario over South America is analysed. This study is based on a set of Regional Climate Models (RCM) from the CORDEX database and their driving Global Climate Models (GCM) from the CMIP5 Project. Historical simulations for 1979–2005 and from the RCP4.5 scenario for 2071–2100 are used. The projected changes in the precipitation behaviour are evaluated in terms of the consistency between the pairs RCM-GCM aiming to explore the added value of RCMs. The agreement between projected changes from RCMs is also evaluated as a measure of confidence of the regional climate change signal. The analysis is carried out for two extended seasons (April to September and October to March). The projected change in the mean precipitation over subtropical latitudes is associated with both changes in the frequency of rainy days and in the intensity of extreme events, while for the tropics the changes are mainly associated with changes in the wet day frequency. The increase of extreme precipitation events over the subtropics arises as a robust signal among models, while for tropical latitudes the dispersion among models is high, which reduces the confidence of the projections. In general, the consistency of the projections in the pairs RCM-GCM and the agreement among models are higher for the low frequency variability patterns during winter, while for summertime a better agreement in the projected changes of the precipitation behaviour at different timescales among RCMs is found.

The possible future changes in temperature over Nigeria were projected in this study. Using Climate Research Unit (CRU) temperature as the reference data, gain ratio (GR), entropy gain (EG), and symmetrical uncertainty (SU) feature selection methods and a multi-criteria decision-making (MCDM) approach were used in selecting the most suitable GCMs for Nigeria from 20 Coupled Model Intercomparison Project phase 5 (CMIP5) global climate models (GCMs). The biases in selected GCMs were corrected using power transformation (PT) method. Multi-model ensembles (MMEs) were generated for the selected GCMs for the different temperature classes’ maximum, average, and minimum for all RCPs. The MMEs were used for the projection of temperatures over the country during 2010–2039, 2040–2069, and 2070–2099. The GCMs HadGEM2-ES, CESM1-CAM5, CSIRO-Mk3.6.0, and MRI-CGCM3 were the best performing in replicating temperature characteristics of the observed temperature in Nigeria. The MME mean projections of bias-corrected (BC) GCMs using PT revealed that there will be an increase in temperature of 4.0 °C at the semi-arid and 5.0 °C at the arid regions during dry and wet seasons respectively under RCP 4.5. In the same regions, the maximum temperature is expected to increase up to 5.5 °C under RCP 8.5 during 2070–2099 in the dry season. In the wet season, temperatures are expected to be higher under RCP 8.5, with an increase of 0.0–4.0 °C in the southern region and 3.0–6.9 °C in the northern region.

This study investigates how subgrid cloud water inhomogeneity within a grid spacing of a general circulation model (GCM) links to the global climate through precipitation processes. The effect of the cloud inhomogeneity on autoconversion rate is incorporated into the GCM as an enhancement factor using a prognostic cloud water probability density function (PDF), which is assumed to be a truncated skewed-triangle distribution based on the total water PDF originally implemented. The PDF assumption and the factor are evaluated against those obtained by global satellite observations and simulated by a global cloud-system-resolving model (GCRM). Results show that the factor implemented exerts latitudinal variations, with higher values at low latitudes, qualitatively consistent with satellite observations and the GCRM. The GCM thus validated for the subgrid cloud inhomogeneity is then used to investigate how the characteristics of the enhancement factor affect global climate through sensitivity experiments with and without the factor incorporated. The latitudinal variation of the factor is found to have a systematic impact that reduces the cloud water and the solar reflection at low latitudes in the manner that helps mitigate the too-reflective cloud bias common among GCMs over the tropical oceans. Due to the limitation of the factor arising from the PDF assumption, however, no significant impact is found in the warm rain formation process. Finally, it is shown that the functional form for the PDF in a GCM is crucial to properly characterize the observed cloud water inhomogeneity and its relationship with precipitation.

Global climate models (GCMs) are developed to simulate past climate and produce projections of climate in future. Their roles in ascertaining regional issues and possible solutions in water resources planning/management are appreciated across the world. However, there is substantial uncertainty in the future projections of GCM(s) for practical and regional implementation which has attracted criticism by the water resources planners. The present paper aims at reviewing the selection of GCMs and focusing on performance indicators, ranking of GCMs and ensembling of GCMs and covering different geographical regions. In addition, this paper also proposes future research directions.

The current research aimed to evaluate the predictive skill of statistically downscaled National Aeronautics Space Administration (NASA) Earth Exchange Global Daily Downscaled Projection (NEX-GDDP) data in simulating the Indian summer monsoon rainfall (ISMR) for the period of 1961–2005 over the individual homogeneous monsoon regions of India (HMRI). For the purpose, five models are selected, as these models (in GCM) have shown better performance in the simulation of ISMR by the researcher. The spatial characteristics and statistical scores (annual cycle, percentage bias, Taylor score, probability distribution function) are used to evaluate the performance of each model in simulating rainfall over land points of individual HMRI. In the spatial analysis, it seems that models of NEX-GDDP can simulate the ISMR, pretty well in comparison to APHRODITE (observation), and show a moderate to significantly high correlation (grid point) over each of the HMRI particularly to core monsoon region, except over few parts of PI. The Taylor statistics suggest that the model CanESM performs very well over the regions of PI, NWI, and WCI. The models MPI-ESM-LR and NorESM perform well in simulating the ISMR over CNI, followed by ACCESS, CanESM, and CCSM4. The models have varying bias in predicting the rainfall; however, ACCESS does perform well and shows the minimum bias (ranges from ~ 1 to ~ 14% only) among others. The models CanESM and NorESM (except over CNI) performed relatively better. The NEX-GDDP models overcome the global climate models (GCMs) in the retrospective simulation of ISMR over the land points of India. It is concluded that the models have good predictability of JJAS rainfall but unable to catch daily rainfall variability.

A multimodel ensemble of general circulation models (GCM) is a popular approach to assess hydrological impacts of climate change at local, regional, and global scales. The traditional multimodel ensemble approach has not considered different uncertainties across GCMs, which can be evaluated from the comparisons of simulations against observations. This study developed a comprehensive index to generate an optimal ensemble for two main climate fields (precipitation and temperature) for the studies of hydrological impacts of climate change over China. The index is established on the skill score of each bias-corrected model and different multimodel combinations using the outputs from phase 5 of the Coupled Model Intercomparison Project (CMIP5). Results show that the optimal ensemble of the nine selected models accurately captures the characteristics of spatial–temporal variabilities of precipitation and temperature over China. We discussed the uncertainty of subset ensembles of ranking models and optimal ensemble based on historical performance. We found that the optimal subset ensemble of nine models has relative smaller uncertainties compared with other subsets. Our proposed framework to postprocess the multimodel ensemble data has a wide range of applications for climate change assessment and impact studies.

Understanding the reliability of global climate models (GCMs) to reproduce the historical surface wind fields is integral part of building robust projections of surface wind-climate, and other wind-dependent geophysical climatic variables. Understanding the skill of atmosphere-only models (AGCM), coupled atmosphere–ocean models (AOGCM) and fully coupled earth system models (ESM) is likewise paramount to assess any systematic model improvements. In this paper, we systematically assess whether surface wind fields obtained from 28 CMIP5 GCMs can represent large-scale spatial patterns and temporal variability of historical surface winds. We show that inter-model uncertainty is typically 2–4 times larger than the uncertainty associated with GCM internal variability, although the latter can be significant within specific regions. We also find that CMIP5 models are typically capable of reliably reproducing large-scale spatial patterns of historical near-surface winds, but considerable uncertainty lies within the CMIP5 ensemble with strong latitudinal dependence. CMIP5 models show limitations in their ability to reliably represent inter-annual and inter-seasonal variability particularly within tropical-cyclone-affected regions. In further analysis, we quantify and intercompare historical wind bias from different types of models with different dynamical cores, based on multiple CMIP5 diagnostic experiments. We find that bias in surface wind fields are largely intrinsic to the atmospheric components of the models, and that the inclusion of carbon-cycle dynamics has insignificant effect on simulated surface winds (at decadal time-scales). Inconsistencies between AGCM and AOGCM simulations are largely driven by errors in sea surface temperatures (SST); though such differences are not statistically significant relative to the inter-model uncertainty within the CMIP5 ensemble. These results show that the dominant source of bias in simulated wind fields lies in the underlying physics of the atmospheric component of the models.

To assess deep convective parameterizations in a variety of GCMs and examine the fast-time-scale convective transition, a set of statistics characterizing the pickup of precipitation as a function of column water vapor (CWV), PDFs and joint PDFs of CWV and precipitation, and the dependence of the moisture–precipitation relation on tropospheric temperature is evaluated using the hourly output of two versions of the GFDL Atmospheric Model, version 4 (AM4), NCAR CAM5 and superparameterized CAM (SPCAM). The 6-hourly output from the MJO Task Force (MJOTF)/GEWEX Atmospheric System Study (GASS) project is also analyzed. Contrasting statistics produced from individual models that primarily differ in representations of moist convection suggest that convective transition statistics can substantially distinguish differences in convective representation and its interaction with the large-scale flow, while models that differ only in spatial–temporal resolution, microphysics, or ocean–atmosphere coupling result in similar statistics. Most of the models simulate some version of the observed sharp increase in precipitation as CWV exceeds a critical value, as well as that convective onset occurs at higher CWV but at lower column RH as temperature increases. While some models quantitatively capture these observed features and associated probability distributions, considerable intermodel spread and departures from observations in various aspects of the precipitation–CWV relationship are noted. For instance, in many of the models, the transition from the low-CWV, nonprecipitating regime to the moist regime for CWV around and above critical is less abrupt than in observations. Additionally, some models overproduce drizzle at low CWV, and some require CWV higher than observed for strong precipitation. For many of the models, it is particularly challenging to simulate the probability distributions of CWV at high temperature.

The driver mechanisms of Dansgaard-Oeschger (DO) events remain uncertain, in part because many climate models do not show similar oscillatory behaviour. Here we present results from glacial simulations of the HadCM3B coupled atmosphere–ocean-vegetation model that show stochastic, quasi-periodical variability on a similar scale to the DO events. This variability is driven by variations in the strength of the Atlantic Meridional Overturning Circulation in response to North Atlantic salinity fluctuations. The mechanism represents a salt oscillator driven by the salinity gradient between the tropics and the Northern North Atlantic. Utilising a full set of model salinity diagnostics, we identify a complex ocean–atmosphere-sea-ice feedback mechanism that maintains this oscillator, driven by the interplay between surface freshwater fluxes (tropical P-E balance and sea-ice), advection, and convection. The key trigger is the extent of the Laurentide ice sheet, which alters atmospheric and ocean circulation patterns, highlighting the sensitivity of the climate system to land-ice extent. This, in addition to the background climate state, pushes the climate beyond a tipping point and into an oscillatory mode on a timescale comparable to the DO events.

Evaluating global climate model (GCM) outputs is essential for accurately simulating future hydrological cycles using hydrological models. The GCM multi-model ensemble (MME) precipitation simulations of the Climate Model Intercomparison Project Phases 5 and 6 (CMIP5 and CMIP6, respectively) were spatially and temporally downscaled according to a multi-site statistical downscaling method for the Hanjiang River Basin (HRB), China. Downscaled precipitation accuracy was assessed using data collected from 14 meteorological stations in the HRB. The spatial performances, temporal performances, and seasonal variations of the downscaled CMIP5-MME and CMIP6-MME were evaluated and compared with observed data from 1970–2005. We found that the multi-site downscaling method accurately downscaled the CMIP5-MME and CMIP6-MME precipitation simulations. The downscaled precipitation of CMIP5-MME and CMIP6-MME captured the spatial pattern, temporal pattern, and seasonal variations, however, precipitation was slightly overestimated in the western and central HRB and precipitation was underestimated in the eastern HRB. The precipitation simulation ability of the downscaled CMIP6-MME relative to the downscaled CMIP5-MME improved because of reduced biases. The downscaled CMIP6-MME better simulated precipitation for most stations compared to the downscaled CMIP5-MME in all seasons except for summer. Both the downscaled CMIP5-MME and CMIP6-MME exhibit poor performance in simulating rainy days in the HRB.

The present study assessed the impact of climate change in the Anandapur catchment of Baitarani River basin, India, using the Soil and Water Assessment Tool (SWAT) hydrological model. The future climatic alterations under two Representative Concentration Pathways (RCPs), i.e. 4.5 and 8.5 scenarios, are quantified by an ensemble of two different CMIP5 models, i.e. CNRM-CM5.0, GFDL-CM3.0. The outcomes of this study reveal that the future rainfall and temperature may experience an increasing trend with gradual shifting of monsoon from mid-June to mid-May. The average annual streamflow experienced the highest increase during the period 2071–2095, whereas the highest average annual evapotranspiration (ET) is observed for the period 2046–2070 under both the RCPs and resulting in comparatively slower groundwater recharge (GWR) over the basin. In order to implement suitable adaptation strategies for a possible flood scenario on the concerned study basin, three critical sub-basins, namely, sub-basin 1, 4, and 5, were identified. Furthermore, the altered streamflow and ET dynamics may result in a significant shifting in the conventional agricultural practice in the coming future time scales. Conclusively, the outcomes of this study have potential implications for policy makers in formulating the policies related to sustainable water resources management in future scenarios.

The paper presents a method for the scale-dependent validation of the spatio-temporal variability in global weather or climate models and for their bias quantification in relation to dynamics. The method provides a relationship between the bias and simulated spatial and temporal variance by a model in comparison with verifying reanalysis data. For the low resolution (T30L8) subset of ERA-20C data, it was found that 80–90 (depending on season) of the global interannual variance is at planetary scales (zonal wavenumbers k = 0−3), and only about 1 of the variance is at scales with k> 7. The reanalysis is used to validate a T30L8 GCM in two configurations, one with the prescribed sea-surface temperature (SST) and another using a slab ocean model. Although the model with the prescribed SST represents the average properties of surface fields well, the interannual variability is underestimated at all scales. Similar to variability, model bias is strongly scale dependent. Biases found in the experiment with the prescribed SST are largely increased in the experiment using a slab ocean, especially in k= 0 , in scales with missing variability and in seasons with poorly simulated energy distribution. The perfect model scenario (a comparison between the GCM coupled to a slab ocean vs. the same model with prescribed SSTs) shows that the representation of the ocean is not critical for synoptic to subsynoptic variability, but essential for capturing the planetary scales. © 2019, Springer-Verlag GmbH Germany, part of Springer Nature.

Anthropogenic climate change is projected to lead to a weakening of the Atlantic meridional overturning circulation (AMOC). One of the mechanisms contributing to this is ice melt leading to a freshening of the North Atlantic Ocean. We use two global climate models to investigate the role of temperature and salinity in the weakening of the AMOC resulting from freshwater forcing. This study finds that freshwater hosing reduces the strength of the AMOC, but in some situations it is not through reduced density from freshening, but a reduction in density from subsurface warming. When the freshwater is mixed down it directly reduces the density of the North Atlantic, weakening the strength of the AMOC. As the AMOC weakens, the mixed layer depth reduces and surface properties are less effectively mixed down. A buoyant surface cap forms, blocking atmospheric fluxes. This leads to the development of a warm anomaly beneath the surface cap, which becomes the primary driver of AMOC weakening. We found that the mean North Atlantic salinity anomaly can be used as a proxy for AMOC weakening because it describes the extent of this surface cap.

The impact on seasonal polar predictability from improved tropical and midlatitude forecasts is explored using a perfect-model experiment and applying a nudging approach in a GCM. We run three sets of 7-month long forecasts: a standard free-running forecast and two nudged forecasts in which atmospheric winds, temperature, and specific humidity (U, V, T, Q) are nudged toward one of the forecast runs from the free ensemble. The two nudged forecasts apply the nudging over different domains: the tropics (30°S–30°N) and the tropics and midlatitudes (55°S–55°N). We find that the tropics have modest impact on forecast skill in the Arctic or Antarctica both for sea ice and the atmosphere that is mainly confined to the North Pacific and Bellingshausen–Amundsen–Ross Seas, whereas the midlatitudes greatly improve Arctic winter and Antarctic year-round forecast skill. Arctic summer forecast skill from May initialization is not strongly improved in the nudged forecasts relative to the free forecast and is thus mostly a “local” problem. In the atmosphere, forecast skill improvement from midlatitude nudging tends to be largest in the polar stratospheres and decreases toward the surface.

Observed past climate data used as input in glacier models are expected to differ among datasets, particularly those for precipitation at high elevations. Differences among observed past climate datasets have not yet been described as a cause of uncertainty in projections of future changes in glacier mass, although uncertainty caused by varying future climate projections among general circulation models (GCMs) has often been discussed. Differences among observed past climate datasets are expected to propagate as uncertainty in future changes in glacier mass due to bias correction of GCMs and calibration of glacier models. We project ensemble future changes in the mass of glaciers in Asia through the year 2100 using a glacier model. A set of 18 combinations of inputs, including two observed past air temperature datasets, three observed past precipitation datasets, and future air temperature and precipitation projections from three GCMs were used. The uncertainty in projected changes in glacier mass was partitioned into three distinct sources: GCM uncertainty, observed past air temperature uncertainty, and observed past-precipitation uncertainty. Our findings indicate that, in addition to the differences in climate projections among GCMs, differences among observed past climate datasets propagate fractional uncertainties of about 15% into projected changes in glacier mass. The fractional uncertainty associated with observed past precipitation was 33–50% that of the observed air temperature. Differences in observed past air temperatures and precipitation did not propagate equally into the ultimate uncertainty of glacier mass projection when ablation was dominant.

The time at which climate change signal can be clearly distinguished from noise is known as time of emergence (ToE) and is typically detected by a general circulation model (GCM) signal-to-noise ratio exceeding a certain threshold. ToE is commonly estimated at large scales from GCMs, although management decisions and adaptation strategies are implemented locally. This paper proposes a methodology to estimate ToE for both precipitation and temperature at local scales (i.e., river basin). The methodology considers local climatic conditions and unbiased GCM projections to estimate ToE by using the statistical power to find when the climate significantly differs from the historical one. The method suggests that ToE for temperature already occurred in three Chilean basins (Limarí, Maipo, and Maule). However, in terms of precipitation, an earlier ToE is clearly identified for the Maule basin, indicating that risk assessment and adaptation measures should be implemented first in this basin.

Global radiative feedbacks have been found to vary in global climate model (GCM) simulations. Atmospheric GCMs (AGCMs) driven with historical patterns of sea surface temperatures (SSTs) and sea ice concentrations produce radiative feedbacks that trend toward more negative values, implying low climate sensitivity, over recent decades. Freely evolving coupled GCMs driven by increasing CO2 produce radiative feedbacks that trend toward more positive values, implying increasing climate sensitivity, in the future. While this time variation in feedbacks has been linked to evolving SST patterns, the role of particular regions has not been quantified. Here, a Green’s function is derived from a suite of simulations within an AGCM (NCAR’s CAM4), allowing an attribution of global feedback changes to surface warming in each region. The results highlight the radiative response to surface warming in ascent regions of the western tropical Pacific as the dominant control on global radiative feedback changes. Historical warming from the 1950s to 2000s preferentially occurred in the western Pacific, yielding a strong global outgoing radiative response at the top of the atmosphere (TOA) and thus a strongly negative global feedback. Long-term warming in coupled GCMs occurs preferentially in tropical descent regions and in high latitudes, where surface warming yields small global TOA radiation change but large global surface air temperature change, and thus a less-negative global feedback. These results illuminate the importance of determining mechanisms of warm pool warming for understanding how feedbacks have varied historically and will evolve in the future.

A regional climate model (RegCM4) is used to project climate change over China in the twenty-first century under the RCP4.5 and RCP8.5 pathways. The driving GCM is CSIRO Mk3.6.0 (hereafter referred to simply as CSIRO), and the simulation (hereafter referred to as CdR) is run at a grid spacing of 25 km. The focus of the present paper is on the changes in mean surface air temperature and precipitation in December–January–February (DJF) and June–July–August (JJA) over China. Validation of the model performances is provided first, followed by a comparison of future changes projected by CSIRO and CdR. Substantial warming in the future is simulated by both models, being more pronounced in DJF compared to JJA, and under RCP8.5 compared to RCP4.5. The warming shows different spatial patterns and, to a less extent, magnitude between CSIRO and CdR. Precipitation change shows a general increase in DJF and a mixture of increase and decrease in JJA. Substantial differences between the two models are found in for precipitation change in JJA. The paper further emphasizes the uncertainties in climate change projection over the region.

The influence of sea surface temperature (SST) on interannual surface wind variability in the tropical Atlantic and Pacific is estimated using sensitivity experiments with the SINTEX-F GCM and the ensemble spread in a nine-member control simulation. Two additional estimates are derived for both SINTEX-F and the ERA-Interim reanalysis using regression analysis and singular value decomposition. All methods yield quite consistent estimates of the fraction of surface wind variability that is determined by SST and therefore potentially predictable. In the equatorial Atlantic, analysis suggests that for the period 1982–2014 approximately 2/3 of surface zonal wind variability in boreal spring and early summer is potentially predictable, while 1/3 is due to noise. Of the predictable component, up to about 35% may be driven from outside the tropical Atlantic, suggesting an important role for remote forcing and a diminished one for local feedbacks. In the northern tropical Atlantic, only 30% of boreal winter variability is predictable, most of which is forced from the Pacific. This suggests a minor role for local coupled air–sea feedbacks. For the equatorial Pacific, the results suggest high predictability throughout the year, most of which is due to local SST, with the tropical Atlantic only playing a minor role in boreal summer. In the tropical Atlantic, atmospheric internal variability is strongly dependent on the presence of deep convection, which, in turn, is related to mean SST. A similar, but weaker, state dependence of internal variability is evident in the tropical Pacific.

The River Subansiri, one of the largest tributaries of the Brahmaputra, makes a significant contribution towards the discharge at its confluence with the Brahmaputra. This study aims to investigate an appropriate model to predict the future flow scenario of the river Subansiri. Two models have been developed. The first model is an artificial neural network (ANN)-based rainfall-runoff model where rainfall has been considered as the input. The future rainfall of the basin is calculated using a multiple non-linear regression-based statistical downscaling technique. The proposed second model is a hybrid model developed using ANN and the Soil Conservation Service (SCS) curve number (CN) method. In this model, both rainfall and land use/land cover have been incorporated as the inputs. The ANN models were run using time series analysis and the method selected is the non-linear autoregressive model with exogenous inputs. Using Sen's slope values, the future trend of rainfall and runoff over the basin have been analyzed. The results showed that the hybrid model outperformed the simple ANN model. The ANN-SCS-based hybrid model has been run for different land use/land cover scenarios to study the future flow scenario of the River Subansiri.

Evapotranspiration is one of the most important climate variables for agricultural and environmental management. We investigate the characteristics of daily crop evapotranspiration (ETo) forecasts up to 16 days ahead derived from GCM outputs using the Penman-Monteith Equation. Furthermore, we propose a post-processing method to calibrate the raw ETo forecasts to observations and improve overall reliability and skill. The post-processing method comprises: (1) the Yeo-Johnson transformation to handle the non-normal distribution of daily ETo, (2) the bi-variate normal distribution to formulate the relationship between transformed raw forecasts and observations, and (3) the Schaake shuffle to establish realistic temporal behaviour in each ensemble member. For three distinct Australian case-study locations, ETo forecasts derived from raw ACCESS-S1 outputs are positively correlated with observations up to 16 days ahead. However, raw forecasts verify more poorly than naive climatology forecasts because of significant bias and poor ensemble spreads. Post-processing the raw ACCESS-S1 ETo improves skill and reliability markedly. Skilful daily ETo forecasts are achieved in the first week, and the skill of cumulative ETo forecasts extends beyond two weeks into the future. The proposed ETo post-processing method could be applied to support operational, multi-week forecasting of short-term ETo in support of the day-to-day decisions of irrigators in Australia and elsewhere.

Understanding the spatiotemporal evolution of extreme temperature events is critical to evaluate their influences in nature. Out of 28 general circulation models (GCMs) in the Coupled Model Inter-comparison Project phase 5, eight GCMs were selected to simulate changes of the six extreme temperature indices (ETIs) in China: warmest daily maximum temperature-TXx, warm days-TX90p, tropical days-TD30, coldest daily minimum temperature-TNn, cool nights-TN10p and frost days-FD0. Statistically downscaled daily maximum and minimum temperatures at 552 sites under representative concentration pathway (RCP) 4.5 and 8.5 scenarios were used. The modified Mann-Kendall method was used to detect the trends and their significances in ETIs. The Morlet wavelet basic function was used to investigate the main and quasi periods of the ETIs over the historical (1961–2000) and future (2001−2100) time stages. The results showed that FD0 and TN10p had decreasing trends, and TNn had increasing trends at more sites than decreasing trends. However, TD30, TX90p and TXx showed consistently increasing trends both during historical and future periods under the RCP4.5 and 8.5 scenarios. More sites had significant trends in the 6 ETIs during 2001–2100. The decreases in FD0 and TN10p reached 31.4 and 27.4 days during 2021–2060 and 42.6 and 29.7 days during 2061–2100, respectively. The increases in TNn, TD30, TX90p, and TXx reached 8.7 °C, 45.3 days, 41.6 days and 5.7 °C over 2021–2060 and 7.2 °C, 65.6 days, 62.4 days and 6.4 °C over 2061–2100, respectively. With respect to the averaged values for 552 sites, the primary periods of FD0, TN10p, TNn, TD30, TX90p and TXx were 8, 8, 8, 3, 3 and 10 years from 1961 to 2017 and 4, 4, 4, 2, 2 and 3 years under the RCP4.5 scenario from 2021 to 2100, respectively, whereas only TN10p and TNn had the 2- and 4-year periods under the RCP8.5 scenario. Differently from the averaged ETIs over China, more sites had longer periods from 1961 to 2017 to 2021–2100 under the RCP 4.5 or RCP 8.5 scenarios. This outcome meant that many sites had smaller fluctuations and less periodicity from 2021 to 2100 than from 1961 to 2017. The spatiotemporal changes in the ETIs may lead to the shifting of cultivation, variety adaptability and production for main crops in different sub-regions or climate zones. This research provides projected characteristics of extreme temperature events across China.

Regional climate models (RCMs) are expected to provide better representations of the climate dynamics because of their higher spatial resolutions. Here, we generated an ensemble of decadal (2006–36) RCM predictions for the area of Israel, which spans a considerable climatic gradient and comprises complex terrain. We used the WRF Model forced by the MIROC5 global climate model (GCM). The ensemble was generated by choosing different combinations of radiation, microphysics, surface layer, and planetary boundary layer parameterizations. The simulation results were compared with meteorological station data for the first simulated decade. For the minimum surface temperature, all the RCM configurations performed better than the driving GCM, while for the maximum surface temperature, only three out of eight configurations improved the predictions. The RCM configurations had higher errors in predicting the precipitation, but four configurations had comparable errors to the GCM. For the next two decades, the ensemble average predicts an increase of 0.51° and 0.40°C decade−1 for the average daily minimum and maximum surface temperatures, respectively. No significant change is predicted in the precipitation. We found that all the parameterizations affect the predictions of the surface temperatures and precipitation [e.g., the CAM radiation scheme simulates colder temperatures than the RRTM for GCMs (RRTMG)] but the PBL and surface layer scheme has the largest effect on the errors. Spectral nudging was found to have a considerable effect on the deviations of the precipitation predicted by the WRF configurations from the predictions of the GCM and a much smaller effect on the surface temperature predictions.

The Madden–Julian oscillation (MJO) typically contains several superclusters and numerous embedded mesoscale convective systems (MCSs). It is hypothesized here that the poorly simulated MJOs in current coarse-resolution global climate models (GCMs) is related to the inadequate treatment of unresolved MCSs. So its parameterization should provide the missing collective effects of MCSs. However, a satisfactory understanding of the upscale impact of MCSs on the MJO is still lacking. A simple two-dimensional multicloud model is used as an idealized GCM with clear deficiencies. Eddy transfer of momentum and temperature by the MCSs, predicted by the mesoscale equatorial synoptic dynamics (MESD) model, is added to this idealized GCM. The upscale impact of westward-moving MCSs promotes eastward propagation of the MJO analog, consistent with the theoretical prediction of the MESD model. Furthermore, the upscale impact of upshear-moving MCSs significantly intensifies the westerly wind burst because of two-way feedback between easterly vertical shear and eddy momentum transfer with low-level eastward momentum forcing. Finally, a basic parameterization of the upscale impact of upshear-moving MCSs modulated by deep heating excess and vertical shear strength significantly improves key features of the MJO analog in the idealized GCM with clear deficiencies. A three-way interaction mechanism between the MJO analog, parameterized upscale impact of MCSs, and background vertical shear is identified.

General circulation models (GCMs) are important tools for the study of climate change, but their resolutions are too coarse for station-scale impact assessments. Statistical and dynamical downscaling methods are widely used to translate the predictions of GCMs to the finer spatial scale and it is important to understand the difference between statistical and dynamical downscaling methods in different climatic zones and time periods. Moreover, statistical downscaling can be used on both GCM and regional climate model (RCM) outputs. In this study, two sets of GCM precipitations were dynamically and statistically downscaled and their performances were evaluated against the observed precipitation from 308 stations distributed throughout the Yellow, Yangtze, and Pearl River basins. These stations have distinct climatic characteristics from the historical period (1961–2000) and future period (2031–2050). Results suggest dynamically downscaled GCM precipitation does not present lower biases when comparing observed site-specific precipitation to GCM outputs, and biases of the initial dynamically downscaled GCM outputs decreased in areas with higher humidity. This demonstrates that statistical downscaling can improve GCM and RCM outputs, and the statistical downscaling method can reproduce local-scale precipitation satisfactorily without dynamical downscaling. However, statistical downscaling reduced spatial regularity of the biases that exist in GCM and RCM outputs between the observations and simulation. Additionally, the spatial discrepancy between statistically downscaled GCM and RCM precipitations was very small. In the future period, discrepancies between statistically downscaled RCM and GCM precipitations in the two climate scenarios were larger than the historical period for all climate zones.

Climate classifications of climate models’ outputs have been used to assess environmental changes but systematic analyses of the differences between models, scenarios and classification methods are scarce. Here, the results of applying the most commonly used climate classifications to the outputs of 47 Global Climate Models (GCM) of different physical parameterizations and varied grid size are presented. The extent and intensity of changes for present climate, three different Representative Pathways Scenarios (RCP26, RCP45 and RCP85) and three increasingly-fine classification methods show that there is a consensus between models, and that climate classifications are indeed useful tools to translate physical climatology variables into environmental changes. The main conclusions are that climate classifications can indeed be used to gauge model performance at several grid sizes and that the classification method does not decisively affects the potential global changes in future climates under increasing greenhouse gas emissions. The analyses also reveal that there are several uncertainties that are not attributable to model grid size or to limitations in the reference datasets but more likely to deficiencies in the physics of the models.

Two GCM ensembles indicate that the record sea surface temperatures during the 2017/18 Tasman Sea marine heatwave were virtually impossible without anthropogenic influence. However, natural variability was important in the atmospheric initiation of the event.

Northwest China is characterized by a high water deficit and regular water resource shortages. These issues have become limiting factors for agricultural and socioeconomic development. Based on a trend-preserving method of bias correction, we calibrated the maximum temperature and minimum temperature in four CMIP5 GCMs (CNRM, IPSL, BCC, and CMCC). Then, we investigated variations in the regional crop water requirement (CWR) in the total growth stages for five main crops (cotton, spring corn, summer corn, spring wheat, and winter wheat) in the past (1961–2005) and future (2006–2100). The results suggest that the MK test yielded insignificant decreasing CWR trends in the total growth stages of cotton (0.10 mm/year), spring corn (0.13 mm/year), and spring wheat (0.05 mm/year) and insignificant increasing trends for summer corn (0.02 mm/year) and winter wheat (0.32 mm/year) historically. In the future period, for the same type of crops (cotton), the CWRs in all scenarios (RCP 2.6, 4.5, and 8.5 scenarios) for all GCMs exhibited significant positive trends; for the same GCM (BCC), the CWRs projected for five major crops in the RCP 4.5 and 8.5 scenarios all exhibited extremely significant MK trends (99%); in addition, the CWRs’ rate increases of the five crops projected in RCP8.5 scenario by BCC exhibited the following order: winter wheat (1.25 mm/year), summer corn (1.15 mm/year), spring corn (1.02 mm/year), cotton (0.97 mm/year), and spring wheat (0.87 mm/year). The maximum CWRs of winter wheat were mainly observed in southeastern Northwest China, while those of the other four crops occurred in southern Xinjiang.

Realistic reproduction of historical extreme precipitation has been challenging for both reanalysis and global climate model (GCM) simulations. This work assessed the fidelities of the combined gridded observational datasets, reanalysis datasets, and GCMs [CMIP5 and the Chinese Academy of Sciences Flexible Global Ocean–Atmospheric Land System Model–Finite-Volume Atmospheric Model, version 2 (FGOALS-f2)] in representing extreme precipitation over East China. The assessment used 552 stations’ rain gauge data as ground truth and focused on the probability distribution function of daily precipitation and spatial structure of extreme precipitation days. The TRMM observation displays similar rainfall intensity–frequency distributions as the stations. However, three combined gridded observational datasets, four reanalysis datasets, and most of the CMIP5 models cannot capture extreme precipitation exceeding 150 mm day−1, and all underestimate extreme precipitation frequency. The observed spatial distribution of extreme precipitation exhibits two maximum centers, located over the lower-middle reach of Yangtze River basin and the deep South China region, respectively. Combined gridded observations and JRA-55 capture these two centers, but ERA-Interim, MERRA, and CFSR and almost all CMIP5 models fail to capture them. The percentage of extreme rainfall in the total rainfall amount is generally underestimated by 25%–75% in all CMIP5 models. Higher-resolution models tend to have better performance, and physical parameterization may be crucial for simulating correct extreme precipitation. The performances are significantly improved in the newly released FGOALS-f2 as a result of increased resolution and a more realistic simulation of moisture and heating profiles. This work pinpoints the common biases in the combined gridded observational datasets and reanalysis datasets and helps to improve models’ simulation of extreme precipitation, which is critically important for reliable projection of future changes in extreme precipitation.

Deeper theoretical understanding of Hadley circulation (HC) width and the mechanisms leading to HC expansion is gained by exploring the response of a zonally symmetric slab ocean aquaplanet general circulation model (GCM) to imposed poleward ocean heat transport (OHT). Poleward OHT causes the subtropical edge of the HC to shift poleward by up to 3° compared to its position in simulations without OHT. This HC widening is interpreted as being driven by a decrease in baroclinicity near the poleward edge of the HC and is divided into three components: a decrease in baroclinicity due to 1) a systematic poleward shift of the intertropical convergence zone (ITCZ) during the seasonal cycle that drives a decrease in the angular momentum of the HC and, consequently, a weakening of the vertical shear of the zonal wind; 2) an increase in subtropical static stability and the vertical extent of the HC, both of which result from OHT’s effect on global-mean temperature; and 3) a relaxation of the meridional sea surface temperature (SST) gradient in the outer tropics and subtropics by OHT. Although the third mechanism contributes the most to the response of HC width to OHT, the contributions from the first two mechanisms each account for up to 20%–30% of the HC response. This work highlights the role of ITCZ position in producing HC expansion and in setting the climatological width of the HC, a role which has been underappreciated. This study indicates a fundamental role for baroclinicity in limiting the poleward extent of the HC.

The heat budget of the Pacific equatorial cold tongue (ECT) is explored using the GFDL-FLOR coupled GCM (the forecast-oriented low ocean resolution version of CM2.5) and ocean reanalyses, leveraging the two-layer framework developed in Part I. Despite FLOR’s relatively weak meridional stirring by tropical instability waves (TIWs), the model maintains a reasonable SST and thermocline depth in the ECT via two compensating biases: 1) enhanced monthly-scale vertical advective cooling below the surface mixed layer (SML), due to overly cyclonic off-equatorial wind stress that acts to cool the equatorial source waters; and 2) an excessive SST contrast between the ECT and off-equator areas, which boosts the equatorward heat transport by TIWs. FLOR’s strong advective cooling at the SML base is compensated by strong downward diffusion of heat out of the SML, which then allows FLOR’s ECT to take up a realistic heat flux from the atmosphere. Correcting FLOR’s climatological SST and wind stress biases via flux adjustment (FA) leads to weaker deep advective cooling of the ECT, which then erodes the upper-ocean thermal stratification, enhances vertical mixing, and excessively deepens the thermocline. FA does strengthen FLOR’s meridional shear of the zonal currents in the east Pacific, but this does not amplify either the simulated TIWs or their equatorward heat transport, likely due to FLOR’s coarse zonal ocean resolution. The analysis suggests that to advance coupled simulations of the ECT, improved winds and surface heat fluxes must go hand in hand with improved subseasonal and parameterized ocean processes. Implications for model development and the tropical Pacific observing system are discussed.

Weather Generators (WGs) are one of the major downscaling tools for assessing regional climate change impacts. However, some deficiencies in the performance of WGs have limited their usage. This paper presents a method for correcting the low-frequency variability (LFV) of precipitation in the Improved Weather Generator (IWG) model. The method is based on bias correction in the monthly precipitation distribution of the generated daily series. The performance of the modified model was tested directly by comparing the statistics of generated and observed weather data for 14 stations, and also indirectly by comparing the characteristics of simulated stream-flows of a basin from the simulations run based on generated and observed weather data. The results showed that the method not only corrected the LFV of precipitation but also improved the reproduction of many other statistics. The provided IWG2 model can serve as a useful tool for the downscaling of General Circulation Models (GCMs) scenarios to assess regional climate change impacts, especially hydrological effects.