Recent  

During the last two years tremendous progress in global data-driven weather models trained on re-analysis data has been made. In this study the Pangu-Weather model is compared to several numerical weather prediction models with and without probabilistic post-processing for temperature and wind speed forecasting. The results confirm that global data-driven models are promising for operational weather forecasting and that post-processing can improve these forecasts considerably.

A novel approach, the Optimal Transport Data Assimilation, is introduced to merge data assimilation and optimal transport concepts. By leveraging optimal transport's displacement interpolation in space, it minimises mislocation errors within data assimilation applied to physical fields, such as water vapour, hydrometeors, chemical species, etc. Its richness and flexibility are showcased through one- and two-dimensional illustrations.

The goal of this paper is to weight several dynamic models in order to improve the representativeness of a system. It is illustrated using a set of versions of an idealized model describing the Atlantic Meridional Overturning Circulation. The low-cost method is based on data-driven forecasts. It enables model performance to be evaluated on their dynamics. Taking into account both model performance and codependency, the derived weights outperform benchmarks in reconstructing a model distribution.

Forecasts have uncertainties. It is thus essential to reduce these uncertainties. Such reduction requires uncertainty quantification, which often means running costly models multiple times. The cost limits the number of model runs and thus the quantification’s accuracy. This study proposes a technique that utilizes users’ knowledge of forecast uncertainties to improve uncertainty quantification. Tests show that this technique improves uncertainty reduction.

We compared two ways to define the phase space of low-dimensional attractors describing the evolution of radar precipitation fields. The first defines the phase space by the domain-scale statistics of precipitation fields, such as their mean, spatial and temporal correlations. The second uses principal component analysis to account for the spatial distribution of precipitation. To represent different climates, radar archives over the United States and the Swiss Alpine region were used.

We focus on the computation of Lyapunov vectors for dynamical systems using partial and noisy observations using data assimilation. We study how the errors in the approximated LVs are implicitly affected by the errors in the estimated trajectory. Our results reveal the difference in sensitivity of the individual vectors in low and high dimensions. Additionally, we find that the subspaces defined by the LVs computed from the approximate trajectory are less sensitive than the individual vectors.

The non-stationary kinetic model of turbulence is used to describe the evolution and structure of the upper turbulent layer with the parameters taken from in situ observations. As an example, we use set of data for three cruises made in different areas of the world ocean. With the given profiles of current shear and buoyancy frequency, the theory yields the results that satisfactorily agree with the measurements of the turbulent dissipation rate.

We train neural networks as denoising diffusion models for state generation in the Lorenz 1963 system and demonstrate that they learn an internal representation of the system. We make use of this learned representation and the pre-trained model in two downstream tasks: surrogate modelling and ensemble generation. For both tasks, the diffusion model can outperform other more common approaches. Thus, we see a potential of representation learning with diffusion models for dynamical systems.

We investigated how machine learning can forecast extreme vegetation responses to weather. Examining four models, no single one stood out as the best, though "echo state networks" showed minor advantages. Our results indicate that while these tools are able to generally model vegetation states, they face challenges under extreme conditions. This underlines the potential of artificial intelligence in ecosystem modeling, also pinpointing areas that need further research.

With improper handling of data missing and offsets, it will affect the accuracy of signal interested. The trend in GNSS belong to the GLOSS is the key to get the absolute sea level. However, it is affected by the periodic signals included. Although adaptive EEMD is capable of extraction periodic signal, the missing and offsets are ignored in the previous work. Meanwhile, the time-varying periodic characteristics derived by adaptive EEMD is more conducive to analyze the driving factors.

In the online seminar series “Perspectives on Climate Sciences: from Historical Developments to Future Frontiers (2020–2021) well-known and established scientists from several fields – including mathematics, physics, climate science and ecology – presented their perspectives on the evolution of climate science and on relevant scientific concepts. In this paper, we first give an overview of the content of the seminar series, and then we introduce the written contributions to this special issue.

Phytoplankton are key in estuaries, forming the ecosystem's base. Despite being washed out by river flow and facing a large range of different salinity, they persist. Our simulation of the Elbe estuary shows growth rates and buoyancy help them retain. Riverbanks and tidal flats offer refuge from strong currents. Our findings emphasize the need for careful ecosystem management in estuaries.

The effect of topography in the MT data along a synthetic model of Roorkee-Gangotri profile.The two correction procedures were used to remove the topography distotion from MT data. The FER, TCR1 and TCR2 in Fig. 3 shows that the both correction procedures are capable to remove the topography effect. The similar TR, TCR1and TCR2 responses are no need for topography correction along Roorkee-Gangotri Profile, because the slope angle is less than one degree.

Identifying causes of specific processes is crucial in order to better understand our climate system. Traditionally, correlation analyses have been used to identify cause-effect relationships in climate studies. However, correlation does not imply causation, which justifies the need to use causal methods. We compare two independent causal methods and show that these are superior to classical correlation analyses. We also find some interesting differences between the two methods.

In this study, different from traditional deterministic optimization algorithms, we implement the sampling method to compute the CNOPs in the realistic and predictive coupled ocean-atmosphere model, which reduces the first-order information to the zeroth-order one avoiding the high-cost computation of the gradient. The numerical performance highlights the importance of stochastic optimization algorithms to compute CNOPs and capture initial optimal precursors.

Breaking internal waves at the edge of the ice cover can essentially enhance melting of ice in the Polar regions. We studied the internal solitary waves transformation on smooth ice surfaces, including those on the ice shelf and glacier outlets, and beneath the ridged underside of the ice. If the tidal flow around the large keels is the source of internal waves, then under conditions of strongly ridged ice the waves excited by the tidal flow are dispersed in the vicinity of their formation.

Auroral kilometric radiation (AKR) is a radio emission emitted by the Earth. Due to the complex mixture of phenomena in the magnetosphere, it is tricky to estimate the time difference between the excitation of two systems. In this study, AKR is compared with indices describing the Earth's system. Time differences between the excitation of AKR and the indices are estimated, using mutual information. AKR feels an enhancement before the aurora, but after more polar latitude features.

We investigate a 1D-energy balance model incorporating a bifurcation parameter to represent the impact of carbon dioxide on Earth's energy budget. We provide conditions for the existence of at least three steady-state solutions, studying the minimum points of a potential functional. Additionally, we characterize the uniqueness of the global minimum point and explore its implications on the bifurcation diagram, with possible applications to reaction-diffusion spatially heterogeneous models.