Search

High‐resolution climate change projections are increasingly necessary to inform climate policy and adaptation planning. Downscaling of global climate models (GCMs) is required to simulate the climate at the spatial scale relevant for local impacts. Here, we dynamically downscaled 15 CMIP6 GCMs to a 10 km resolution over Australia using the Conformal Cubic Atmospheric model (CCAM), creating the largest ensemble of high‐resolution downscaled CMIP6 projections for Australia. We compared the host CMIP6 models and downscaled simulations to the Australian Gridded Climate Data (AGCD) observational data and evaluated performance using the Kling‐Gupta efficiency and Perkins skill score. Downscaling improved performance over host GCMs for seasonal temperature and precipitation (10% and 43% respectively), and for annual cycles of temperature and precipitation (6% and 13% respectively). Downscaling also improved the fraction of dry days, reducing the bias for too many low‐rain days. The largest improvements were found in climate extremes, with enhancements to extreme minimum temperatures in all seasons varying from 142% to 201%, and to extreme precipitation of 52% in Austral winter and 47% in summer. The ensemble average integrated skill score improved by 16%. Temperature and precipitation biases were reduced in mountainous and coastal areas. CCAM downscaling outperformed host CMIP6 GCMs at multiple spatial scales and regions—continental Australia, Australian IPCC regions and Queensland's regions—with integrated added value ranging from 9% to 150% and higher over densely populated regions more exposed to climate impacts. This data set will be a valuable resource for understanding future climate changes in Australia. Plain Language Summary: High‐resolution climate models are used for producing climate change projections for assessing regional and local climate change impacts and for adaptation and policy formulation. We completed 15 high‐resolution climate simulations using the Conformal Cubic Atmospheric Model at a 10 km spatial resolution over the Australian continent and surrounds based on global climate models (GCMs) used in the Intergovernmental Panel on Climate Change Assessment Report 6. We evaluated the new high‐resolution data set by comparing it to observations and to their host global model across several regions within Australia. The high‐resolution simulations improved the representation of topography, seasonal temperature and annual cycles of temperature, precipitation, as well as the number of dry days, extreme precipitation and extreme minimum temperatures. The high‐resolution projections show improvement compared to GCMs consistently across spatial scales and regions, being particularly high (150%) over urban areas with high exposure to impacts. This new 10 km data set is the highest resolution, and largest ensemble of the latest climate projections for the Australian continent and surrounds and will be a valuable tool for evaluating future climate change impacts. Key Points: We dynamically downscaled 15 CMIP6 global climate model simulations over Australia to a 10 km spatial resolution using the Conformal Cubic Atmospheric modelTwo new assessment metrics are proposed to assess model performance and added value of downscalingThe integrated added value of downscaling can be as high as 150% over highly populated areas [ABSTRACT FROM AUTHOR]

Bias correction is a vital technique to correct the climate model outputs for regional studies. Prior bias correction methods such as quantile mapping (QM) may be problematic in correcting extreme values. Herein we proposed a novel bias correction method, that is, the piecewise‐quantile mapping (PQM), by combing piecewise mapping with QM, which implements bias corrections on both extreme and nonextreme data. We compared the PQM method and other six commonly used methods in correcting daily precipitation in the Mekong River Basin (MRB) and the Yangtze River Basin (YRB) from two global climate models (GCMs) (i.e., EC‐Earth3 and MPI‐ESM1‐2‐HR) in the Coupled Model Intercomparison Project Phase 6 (CMIP6). Our results show that the PQM perform better in bias correction pertaining to both the precipitation percentiles and most of the extremes indices. For example, the mean absolute error of R20mm (the number of days when precipitation ≥20 mm) is significantly reduced from 21.64 to 9.03 days and from 10.96 to 5.71 days over MRB and YRB, respectively. In addition, the PQM could capture the spatial distributions of extremes indices reasonably well. The PQM developed in this study provides more accurate bias correction of the GCMs outputs, which will reduce the uncertainty in the subsequent analyses, such as climate change impacts on hydrological and biogeochemical cycles, particularly under extreme conditions. [ABSTRACT FROM AUTHOR]

We analyze projected changes in climate extremes (extreme temperatures and heavy precipitation) in the multimodel ensembles of the fifth and sixth Coupled Model Intercomparison Projects (CMIP5 and CMIP6). The results reveal close similarity between both ensembles in the regional climate sensitivity of the projected multimodel mean changes in climate extremes, that is, their projected changes as a function of global warming. This stands in contrast to widely reported divergences in global (transient and equilibrium) climate sensitivity in the two multimodel ensembles. Some exceptions include higher warming in the South America monsoon region, lower warming in Southern Asia and Central Africa, and higher increases in heavy precipitation in Western Africa and the Sahel region in the CMIP6 ensemble. The multimodel spread in regional climate sensitivity is found to be large in both ensembles. In particular, it contributes more to intermodel spread in projected regional climate extremes compared with the intermodel spread in global climate sensitivity in CMIP6. Our results highlight the need to consider regional climate sensitivity as a distinct feature of Earth system models and a key determinant of projected regional impacts, which is largely independent of the models' response in global climate sensitivity. Plain Language Summary: Many articles analyze and compare global climate sensitivity in climate models, that is, how their global warming differs at a given level of CO2 concentrations. However, global warming is only one quantity affecting impacts. To assess human‐ and ecosystem‐relevant impacts, it is essential to evaluate the regional climate sensitivity of climate models, that is, how their regional climate features differ at a given level of global warming. We analyze here regional climate sensitivity in the new multimodel ensemble that will underlie the conclusions of the Sixth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC). This ensemble of model projections is called the "Sixth Coupled Model Intercomparison Project" or CMIP6. We find that differences in regional climate sensitivity between models in CMIP6 often contribute more to the uncertainty of regional extremes projections than the uncertainty in global mean warming between models. Overall, the regional climate sensitivity features in the CMIP6 models' projections ensemble are very similar to those of the prior ensemble (CMIP5), although the model ensembles have been highlighted to differ in their global climate sensitivity over the 21st century. Key Points: Changes in climate extremes as a function of global warming are quasilinear and determine a "regional climate sensitivity" in CMIP5 and CMIP6The regional climate sensitivity of climate extremes is found to be very similar in CMIP5 and CMIP6, unlike global climate sensitivityModel spread in regional climate sensitivity in CMIP6 contributes more to uncertainty of projected extremes than global climate sensitivity [ABSTRACT FROM AUTHOR]