The 2019–2020 summer wildfire event on the east coast of Australia was a series of major wildfires occurring from November 2019 to end of January 2020 across the states of Queensland, New South Wales (NSW), Victoria and South Australia. The wildfires were unprecedent in scope and the extensive character of the wildfires caused smoke pollutants to be transported not only to New Zealand, but also across the Pacific Ocean to South America. At the peak of the wildfires, smoke plumes were injected into the stratosphere at a height of up to 25 km and hence transported across the globe. The meteorological and air quality Weather Research and Forecasting with Chemistry (WRF-Chem) model is used together with the air quality monitoring data collected during the bushfire period and remote sensing data from the Moderate Resolution Imaging Spectroradiometer (MODIS) and Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation (CALIPSO) satellites to determine the extent of the wildfires, the pollutant transport and their impacts on air quality and health of the exposed population in NSW. The results showed that the WRF-Chem model using Fire Emission Inventory (FINN) from National Center for Atmospheric Research (NCAR) to simulate the dispersion and transport of pollutants from wildfires predicted the daily concentration of PM_2.5 having the correlation (R²) and index of agreement (IOA) from 0.6 to 0.75 and 0.61 to 0.86, respectively, when compared with the ground-based data. The impact on health endpoints such as mortality and respiratory and cardiovascular diseases hospitalizations across the modelling domain was then estimated. The estimated health impact on each of the Australian Bureau of Statistics (ABS) census districts (SA4) of New South Wales was calculated based on epidemiological assumptions of the impact function and incidence rate data from the 2016 ABS and NSW Department of Health statistical health records. Summing up all SA4 census district results over NSW, we estimated that there were 247 (CI: 89, 409) premature deaths, 437 (CI: 81, 984) cardiovascular diseases hospitalizations and 1535 (CI: 493, 2087) respiratory diseases hospitalizations in NSW over the period from 1 November 2019 to 8 January 2020. The results are comparable with a previous study based only on observation data, but the results in this study provide much more spatially and temporally detailed data with regard to the health impact from the summer 2019–2020 wildfires.

Convection is often assumed to be controlled by the simultaneous environmental fields. But to what extent does it also remember its past behavior? This study proposes a new framework in which the memory of previous convective-scale behavior, “microstate memory,” is distinguished from macrostate memory, and conducts numerical experiments to reveal these memory types. A suite of idealized, cloud-resolving radiative–convective equilibrium simulations in a 200-km square domain is performed with the Weather Research and Forecasting (WRF) Model. Three deep convective cases are analyzed: unorganized, organized by low-level wind shear, and self-aggregated. The systematic responses to sudden horizontal homogenization of various fields, in various atmospheric layers, designed to eliminate their specific microstructure, are compared in terms of precipitation change and time of recovery to equilibrium. Results imply a substantial role for microstate memory. Across organization types, microstructure in water vapor and temperature has a larger and longer-lasting effect on convection than in winds or hydrometeors. Microstructure in the subcloud layer or the shallow cloud layer has more impact than in the free troposphere. The recovery time scale dramatically increases from unorganized (2–3 h) to organized cases (24 h or more). Longer-time-scale adjustments also occur and appear to involve both small-scale structures and domain-mean fields. These results indicate that most convective microstate memory is stored in low-level thermodynamic structures, potentially involving cold pools and hot thermals. This memory appears strongly enhanced by convective organization. Implications of these results for parameterizing convection are discussed.

This study explores the role of the stratiform cloud scheme in the inter‐model spread of cloud feedback. Six diagnostic cloud schemes used in various CMIP (Coupled Model Intercomparison Experiment] climate models are implemented (at low and midlevels) into two testbed climate models, and the impacts on cloud feedback are investigated. Results suggest that the choice of stratiform cloud scheme may contribute up to roughly half of the intermodel spread of cloud radiative responses in stratocumulus (Sc) regions, and may determine or favor a given sign of the feedback there. Cloud schemes assuming a probability density function for total water content consistently predict a positive feedback in Sc regions in our experiments. A large negative feedback in Sc regions is obtained only with schemes that consider variables other than relative humidity (e.g., stability). The stratiform cloud scheme also significantly affects cloud feedback at the scale of the tropics and at global scale. Results are slightly less consistent for tropical means, likely indicating coupling with other boundary layer processes such as convective mixing.

The coupled Conformal Cubic Atmospheric Model (CCAM) and Chemical Transport Model (CTM) (CCAM-CTM) was undertaken with eleven emission scenarios segregated from the 2008 New South Wales Greater Metropolitan Region (NSW GMR) Air Emission Inventory to predict major source contributions to ambient PM_2.5 and exposure in the NSW GMR. Model results illustrate that populated areas in the NSW GMR are characterised with annual average PM_2.5 of 6–7 µg/m³, while natural sources including biogenic emissions, sea salt and wind-blown dust contribute 2–4 µg/m³ to it. Summer and winter regional average PM_2.5 ranges from 5.2–6.1 µg/m³ and 3.7–7.7 µg/m³ across Sydney East, Sydney Northwest, Sydney Southwest, Illawarra and Newcastle regions. Secondary inorganic aerosols (particulate nitrate, sulphate and ammonium) and sodium account for up to 23% and 18% of total PM_2.5 mass in both summer and winter. The increase in elemental carbon (EC) mass from summer to winter is found across all regions but particularly remarkable in the Sydney East region. Among human-made sources, “wood heaters” is the first or second major source contributing to total PM_2.5 and EC mass across Sydney in winter. “On-road mobile vehicles” is the top contributor to EC mass across regions, and it also has significant contributions to total PM_2.5 mass, particulate nitrate and sulphate mass in the Sydney East region. “Power stations” is identified to be the third major contributor to the summer total PM_2.5 mass across regions, and the first or second contributor to sulphate and ammonium mass in both summer and winter. “Non-road diesel and marine” plays a relatively important role in EC mass across regions except Illawarra. “Industry” is identified to be the first or second major contributor to sulphate and ammonium mass, and the second or third major contributor to total PM_2.5 mass across regions. By multiplying modelled predictions with Australian Bureau of Statistics 1-km resolution gridded population data, the natural and human-made sources are found to contribute 60% (3.55 µg/m³) and 40% (2.41 µg/m³) to the population-weighted annual average PM_2.5 (5.96 µg/m³). Major source groups “wood heaters”, “industry”, “on-road motor vehicles”, “power stations” and “non-road diesel and marine” accounts for 31%, 26%, 19%, 17% and 6% of the total human-made sources contribution, respectively. The results in this study enhance the quantitative understanding of major source contributions to ambient PM_2.5 and its major chemical components. A greater understanding of the contribution of the major sources to PM_2.5 exposures is the basis for air quality management interventions aiming to deliver improved public health outcomes.

The fluctuation–dissipation theorem (FDT) has been proposed as a method of calculating the mean response of the atmosphere to small external perturbations. This paper explores the application of the theory under time and space constraints that approximate realistic conditions. To date, most applications of the theory in the climate context used univariate, low-dimensional-state representations of the climate system and an arbitrarily long sample size. The authors explore high-dimensional multivariate FDT operators and the lower bounds of sample size needed to construct skillful operators. It is shown that the skill of the operator depends on the selection of variables and features representing the climate system and that these features change once memory (slab ocean) is added to the system. In addition, it is found that the FDT operator has skill in estimating the response to realistic sea surface temperature (SST) patterns, such as El Niño–Southern Oscillation (ENSO), despite the fact that these patterns were not part of the data used to produce the operator. The response of clouds is also studied; for variables that represent cloud properties, the decrease in skill in relation to decrease in sample size still maintains the key features of the response.

Climate models are hindered by the need to conceptualize and then parameterize complex physical processes that are not explicitly numerically resolved and for which no rigorous theory exists. Machine learning and artificial intelligence methods (ML and AI) offer a promising paradigm that can augment or replace the traditional parameterized approach with models trained on empirical process data. We offer a flexible and efficient plugin, TorchClim, that facilitates the insertion of ML and AI physics surrogates into the climate model to create hybrid models. A reference implementation is presented for the Community Earth System Model (CESM), where moist physics and radiation parameterizations of the Community Atmospheric Model (CAM) are replaced with such a surrogate. We present a set of best-practice principles for doing this with minimal changes to the general circulation model (GCM), exposing the surrogate model as any other parameterization module, and discuss how to accommodate the requirements of physics surrogates such as the need to avoid unphysical values and supply information needed by other GCM components. We show that a deep-neural-network surrogate trained on data from CAM itself can produce a model that reproduces the climate and variability in the original model, although with some biases. The efficiency and flexibility of this approach open up new possibilities for using physics surrogates trained on offline data to improve climate model performance, better understand model physical processes, and flexibly incorporate new processes into climate models.