An integral and expensive part of coupled climate model simulations is the gas-phase chemistry which gives rise to hundreds of coupled differential equations. We propose a method which improves the convergence and robustness properties of commonly used Newton–Raphson solvers. The method is flexible and can be appended to most algorithms. The approach can be useful for a broader community of computational scientists whose interests lie in solving systems with intensive interactive chemistry.

The TransCom inter-comparison project regularly carries out studies to quantify errors in simulated atmospheric transport. This paper presents the first results of an age of air (AoA) inter-comparison of six global transport models. Following a protocol, six models simulated five tracers from which atmospheric transport times can easily be deduced. Results highlight that inter-model differences associated with atmospheric transport are still large and require further analysis.

Global sensitivity analysis (GSA) identifies which parameters of a model most affect its output. We performed GSA using statistical emulators as surrogates of two slow-running atmospheric chemistry transport models. Due to the high dimension of the model outputs, we considered two alternative methods: one that reduced the output dimension and one that did not require an emulator. The alternative methods accurately performed the GSA but were significantly faster than the emulator-only method.

Large uncertainty is inherent in model predictions due to imperfect knowledge of how to describe the processes that a model is intended to represent. Yet methods to quantify and evaluate this model hypothesis uncertainty are limited. To address this, the multi-assumption architecture and testbed (MAAT) automates the generation of all possible models by combining multiple representations of multiple processes. MAAT provides a formal framework for quantification of model hypothesis uncertainty.

We document the latest version of the shared UK global configuration of the NEMO ocean model. This configuration will be used as part of the climate models for the UK contribution to the IPCC 6th Assessment Report. 30-year integrations forced with atmospheric forcing show that the new configurations have an improved simulation in the Southern Ocean with the near-surface temperatures and salinities and the sea ice all matching the observations more closely.

Summer precipitation over China in the MetUM reaches twice its observed values. Increasing the horizontal resolution of the model and adding air–sea coupling have little effect on these biases. Nevertheless, MetUM correctly simulates spatial patterns of temporally coherent precipitation and the associated large-scale processes. This suggests that the model may provide useful predictions of summer intraseasonal variability despite the substantial biases in overall intraseasonal variance.

Atmospheric aerosols interact with our climate system and have adverse health effects. Nevertheless, these particles are a source of uncertainty in climate projections and the formation process of secondary aerosols formed by organic gas-phase precursors is particularly not fully understood. In order to gain a deeper understanding of secondary organic aerosol formation, this model system explicitly represents gas-phase and aerosol formation processes. Finally, this allows for process discussion.

Clouds form when water condenses onto particles in the atmosphere and the size and chemical composition of these particles can have a large influence over how much water condenses and the subsequent formation of cloud. Additional gases exist in the atmosphere that can condense onto the aerosol particles and change their composition. We present a fast and efficient method of calculating the effect of atmospheric gases on the formation of cloud that can be used in climate and weather models.

Peatlands are widely present in boreal regions and contain large carbon stocks due to their hydrologic properties and high water content. We have enhanced the global land surface model ORCHIDEE by introducing boreal peatlands. These are considered as a new type of vegetation in the model, with specific hydrological properties for peat soil. In this paper, we focus on the representation of the hydrology of northern peatlands and on the evaluation of the hydrological impact of this implementation.

Well-performing, easy-to-use ocean models are a central ingredient to further the understanding of our Earth and climate. Veros, the versatile ocean simulator, is the first full-blown ocean model entirely written in the high-level programming language Python. It is considerably more approachable than traditional Fortran models and leverages modern best practices; at the same time, thanks to the Bohrium framework, Veros is about half as fast as a reference implementation in Fortran 90.

We present a transparent method for evaluating how changes to the MIT Earth System Model impact its response to anthropogenic and natural forcings. We tested the effects that changes to both model components and forcings have on the estimates of model parameters that agree with historical observations. Overall, changes to model forcings are more important than the new components, while the long-term model response is unchanged. The methodology serves as a guide for documenting model development.

Global overviews of upcoming flood and drought events are key for many applications from agriculture to disaster risk reduction. Seasonal forecasts are designed to provide early indications of such events weeks or even months in advance. This paper introduces GloFAS-Seasonal, the first operational global-scale seasonal hydro-meteorological forecasting system producing openly available forecasts of high and low river flow out to 4 months ahead.

The authors discuss the implementation and performance of a simple one-dimensional sea ice scheme within the ALADIN–HIRLAM numerical weather prediction system. This scheme is developed to replace the prescribed sea ice surface temperature that was used before. Numerical experiments over the large ice-covered areas of the European Arctic show that new scheme helps to improve the modelled 2 m temperature in regions that are influenced by sea ice, especially when the forecast is longer than 12 h.

A new module, ORACLE 2-D, that calculates the concentrations of surrogate organic species in two-dimensional space defined by volatility and oxygen-to-carbon ratio has been developed and evaluated. ORACLE 2-D uses a simple photochemical aging scheme that efficiently simulates the net effects of fragmentation and functionalization. ORACLE 2-D can be used to compute the ability of organic particles to act as cloud condensation nuclei and serves as a tool to quantify their climatic impact.

Chemistry is an important component of the atmosphere that describes many important physical processes. However, atmospheric chemical mechanisms include hundreds of species and reactions, posing a significant computational load. In this work, we use a powerful reduction method in order to develop a computationally faster chemical mechanism from a detailed mechanism. This enables accelerated simulations, which can be used to examine a wider range of processes in increased detail.

The impacts of collective algorithms, interconnect topologies, and routing algorithms on the performance and scalability of transpositions, halo exchanges, and allreduce operations are significant. The performance of communications can be improved when the congestion is mitigated by a proper configuration of the topology and a congestion-balanced routing algorithm. The scalability of the spectral models could be acceptable for exascale supercomputers.

This paper gives a detailed description of the model theory of the obstacle-resolving microscale meteorological model MITRAS version 2. Detailed descriptions of the model equations and their formulations and approximations are presented. Also, detailed parameterizations of buildings, wind turbines, and vegetation in the model are introduced. Some example applications of the model are shown to demonstrate the model capacities and potential.

Developing computer models in the earth-system domain is a complex and expensive process that can have a duration measured in years. The supercomputers required to run these models, however, are evolving fast with a proliferation of technologies and associated programming models. As a result there is a need that models be "performance portable" between different supercomputers. This paper investigates a way of doing this through a separation of the concerns of performance and natural science.

The core of every climate model is the solution of the surface energy balance. Numerical approaches are mandatory to calculate the land's response to solar input. However, different numerical approaches should not affect the physical results. Here we develop a physical approach that determines how the available energy is divided into radiative and heat fluxes. A key element of this scheme is the inclusion of different types of heat storages in the canopy layer.

Variable Infiltration Capacity (VIC) is a widely used hydrologic model. This paper documents the development of VIC version 5, which includes a reconfiguration of the model source code to support a wider range of modeling applications. It also represents a significant step forward for the VIC user community in terms of support for a range of modeling applications, reproducibility, and scientific robustness.

We present and validate BrAHMs (BAsal Hydrology Model): a new physically based basal hydrology model, which captures the two main types of subglacial drainage systems (high-pressure distributed systems and low-pressure channelized systems). BrAHMs is designed for continental glacial cycle scale contexts, for which computational speed is essential. This speed is accomplished, in part, by numerical methods novel to basal hydrology contexts.

We have implemented a regional, high-resolution, and computationally attractive carbon dioxide data assimilation system. This system, named CTDAS-Lagrange, is capable of simultaneously optimizing terrestrial biosphere fluxes and the lateral boundary conditions. The CTDAS-Lagrange system can be easily extended to assimilate an additional tracer, e.g., carbonyl sulfide (COS or OCS), for regional estimates of both net and gross carbon fluxes.