With IntelliO3-ts v1.0, we present an artificial neural network as a new forecasting model for daily aggregated near-surface ozone concentrations with a lead time of up to 4 d. We used measurement and reanalysis data from more than 300 German monitoring stations to train, fine tune, and test the model. We show that the model outperforms standard reference models like persistence models and demonstrate that IntelliO3-ts outperforms climatological reference models for the first 2 d.

A first-of-its-kind scheme has been developed for assimilating Ozone Monitoring Instrument (OMI) aerosol index (AI) measurements into the Naval Aerosol Analysis and Predictive System. Improvements in model simulations demonstrate the utility of OMI AI data assimilation for improving the accuracy of aerosol model analysis over cloudy regions and bright surfaces. This study can be considered one of the first attempts at direct radiance assimilation in the UV spectrum for aerosol analyses.

We propose an experimental design of a suite of multi-year, short-term hindcasts and compare them with corresponding observations or measurements for periods based on different weather and climate phenomena. This atypical way of evaluating model performance is particularly useful and beneficial, as these hindcasts can give scientists a robust picture of modeled precipitation, and cloud and radiation processes from their diurnal variation to year-to-year variability.

This paper presents a new interpolator useful for geophysics applications. It can explore N-dimensional meshes, grids or look-up tables. The code accepts irregular but structured grids. Written in Fortran, it is easy to implement in existing codes and very fast and portable. We have compared it with a Python library. Python is convenient but suffers from portability and is sometimes not optimized enough. As an application case, this method is applied to atmospheric sciences.

Detecting extreme weather events is a crucial step in understanding how they change due to climate change. Deep learning (DL) is remarkable at pattern recognition; however, it works best only when labeled datasets are available. We create ClimateNet – an expert-labeled curated dataset – to train a DL model for detecting weather events and predicting changes in extreme precipitation. This work paves the way for DL-based automated, high-fidelity, and highly precise analytics of climate data.

Temperature is a controller of metabolic processes and therefore also a controller of the ocean's biological carbon pump (BCP). We calibrate a temperature-dependent version of the BCP in the cGENIE Earth system model. Since the pre-industrial period, warming has intensified near-surface nutrient recycling, supporting production and largely offsetting stratification-induced surface nutrient limitation. But at the same time less carbon that sinks out of the surface then reaches the deep ocean.

This paper presents updated European prediction algorithms (euptf2) to compute soil hydraulic parameters from easily available soil properties. The new algorithms lead to significantly better predictions and provide a built-in prediction uncertainty computation. The influence of predictor variables on predicted soil hydraulic properties is explored and practical guidance on how to use the derived PTFs is provided. A website and an R package facilitate easy application of the updated predictions.

A cloud macrophysics scheme utilizing grid-mean hydrometeor information is developed and evaluated for climate models. The GFS–TaiESM–Sundqvist (GTS) scheme can simulate variations of cloud fraction associated with relative humidity (RH) in a more consistent way than the default scheme of CAM5.3. Through better cloud–RH distributions, the GTS scheme helps to better represent cloud fraction, cloud radiative forcing, and thermodynamic-related climatic fields in climate simulations.

The spin-up in GRAPES_GFS, under different initial fields, goes through a dramatic adjustment in the first half-hour of integration and slow dynamic and thermal adjustments afterwards. It lasts for at least 6 h, with model adjustment gradually completed from lower to upper layers in the model. Thus, the forecast results, at least in the first 6 h, should be avoided when used. In addition, the spin-up process should repeat when the model simulation is interrupted.

Few studies have investigated the effects of initial conditions on the simulation or prediction of PM_2.5 concentrations. Here, sensitivity experiments are used to explore the effects of three initial mechanisms (clean, restart, and continuous) and emissions in Xi’an in December 2016. According to this work, if the restart mechanism cannot be used due to computing resource and storage space limitations when forecasting PM_2.5 concentrations, a spin-up time of at least 27 h is needed.

The adoption of snowpack models in support of avalanche forecasting has been limited. To promote their operational application, we present a numerical method for processing multivariate snow stratigraphy profiles of mixed data types. Our algorithm enables applications like dynamical grouping and summarizing of model simulations, model evaluation, and data assimilation. By emulating the human analysis process, our approach will allow forecasters to familiarly interact with snowpack simulations.

A new method to estimate the sulfate aerosol hygroscopicity parameter (κSO₄) is suggested that can consider κSO₄ for two different sulfate species instead of prescribing a single κSO₄ value, as in most previous studies. The new method simulates more realistic cloud droplet concentrations and, thus, a more realistic cloud albedo effect than the original method. The new method is simple and readily applicable to modeling studies investigating sulfate aerosols’ effect in aerosol–cloud interactions.

FORTE 2.0 is a flexible coupled atmosphere–ocean general circulation model that can be run on modest hardware. We present two 2000-year simulations which show that FORTE 2.0 is capable of producing a stable climate. Earlier versions of FORTE were used for a wide range of studies, ranging from aquaplanet configurations to investigating the cold European winters of 2009–2010. This paper introduces the updated model for which the code and configuration are now publicly available.

A biogeochemical model of planktonic food webs including a carbonate balance module is applied in the Bay of Marseille (France) to represent the carbon marine cycle expected to change in the future owing to significant increases in anthropogenic emissions of CO₂. The model correctly simulates the ranges and seasonal dynamics of most variables of the carbonate system (pH). This study shows that external physical forcings have an important impact on the carbonate equilibrium in this coastal area.

A synthetic inflow turbulence generator was implemented in the idealised Weather Research and Forecasting large eddy simulation. The inflow case yielded a mean velocity profile and second-moment profiles that agreed well with those generated using periodic boundary conditions, after a short adjustment distance. This implementation can be extended to a multi-scale seamless nesting simulation from a meso-scale domain with a kilometre-scale resolution to LES domains with metre-scale resolutions.

Climate change is a fact and adaptation a necessity. The Economics of Climate Adaptation methodology provides a framework to integrate risk and reward perspectives of different stakeholders, underpinned by the CLIMADA impact modelling platform. This extended version of CLIMADA enables risk assessment and options appraisal in a modular form and occasionally bespoke fashion yet with high reusability of functionalities to foster usage in interdisciplinary studies and international collaboration.

Simple climate models are frequently used in research and decision-making communities because of their tractability and low computational cost. Simple climate models are diverse, including highly idealized and process-based models. Here we present a hybrid approach that combines the strength of two types of simple climate models in a flexible framework. This hybrid approach has provided insights into the climate system and opens an avenue for investigating radiative forcing uncertainties.

Resetting of non-significant figures (precision trimming) enables efficient data compression and helps to avoid excessive use of storage space and network bandwidth while having well-constrained distortion to the data. The paper analyses accuracy losses and artifacts caused by trimming methods and by the widely used linear packing method. The paper presents several methods with implementation, evaluation, and illustrations and includes subroutines directly usable in geoscientific models.