Large-scale Climate Models Research Articles

Current progress in subseasonal-to-decadal prediction based on machine learning

The application of machine learning (ML) techniques to climate science has received significant attention, particularly in the field of climate predictions, ranging from sub-seasonal to decadal time scales. This paper reviews recent progress of ML techniques employed in climate phenomena prediction and the enhancement of dynamic forecast models, which provide valuable insights into the great potentials of ML techniques to improve climate prediction capabilities with reduced computational time and resource consumption. This paper also discusses several major challenges in the application of ML to climate prediction, including the scarcity of datasets, physical inconsistency, and lack of model transparency and interpretability. Additionally, this paper sheds light on how climate change impacts ML model training and prediction, and explores three key areas with potential breakthroughs: large-scale climate models, knowledge discovery driven by ML, and hybrid dynamical-statistical models, underscoring the important role of the integration of “ML and dynamical models” in building a bridge between the artificial intelligence and climate science.

Storm surge modelling along European coastlines: The effect of the spatio-temporal resolution of the atmospheric forcing

The spatio-temporal resolution of atmospheric forcing plays a key role in the accuracy of simulated storm surges with hydrodynamic numerical models. Here, we generate five hydrodynamic hindcasts of coastal storm surges along the European Atlantic and the Mediterranean Sea coasts, forced with atmospheric fields of varying temporal (hourly and daily) and spatial (0.25° to 2°) resolution since 1940. Our results, that are validated with insitu tide gauge observations, show that storm surges obtained with daily forcing underestimate the magnitude of coastal extreme sea level events by up to 50% compared to hourly simulations and observations. Nevertheless, low-resolution simulations capture the temporal variability of storm surges, including strong episodes. Furthermore, taking advantage of the consistent set of coastal storm surge hindcasts, we demonstrate that storm surges forced with daily mean atmospheric fields, when bias corrected via quantile mapping, provide accurate values of daily maxima as calculated by a high-resolution hindcast. This transformation paves the way to obtain daily maxima storm surge estimates from low-resolution atmospheric fields, as those typically provided by large-scale and global climate models, at a lower computational cost.

Hierarchical Autoencoder-Based Lossy Compression for Large-Scale High-Resolution Scientific Data

Lossy compression has become essential an important technique to reduce data size in many domains. This type of compression is especially valuable for large-scale scientific data, whose size ranges up to several petabytes. Although Autoencoder-based models have been successfully leveraged to compress images and videos, such neural networks have not widely gained attention in the scientific data domain. Our work presents a neural network that not only significantly compresses large-scale scientific data, but also maintains high reconstruction quality. The proposed model is tested with scientific benchmark data available publicly and applied to a large-scale high-resolution climate modeling data set. Our model achieves a compression ratio of 140 on several benchmark data sets without compromising the reconstruction quality. 2D simulation data from the High-Resolution Community Earth System Model (CESM) Version 1.3 over 500 years are also being compressed with a compression ratio of 200 while the reconstruction error is negligible for scientific analysis.

A novel methodology for calculating the aerosol direct radiative effect using a non-uniform five-dimensional lookup table

ABSTRACT The quantification of aerosol direct radiative effects (ADREs) under clear sky conditions is essential for understanding the influence of aerosols on climate. This paper introduces a novel approach, the non-uniform five-dimensional lookup table (NU-5D-LUT) method, for calculating the ADREs under clear sky conditions. The NU-5D-LUT method considers key parameters such as aerosol optical thickness, single scattering albedo, asymmetry factor, surface albedo, and solar zenith angle. Validation against the SBDART model shows strong agreement, with correlation coefficients of 0.97 and 0.99 for TOA and BOA ADREs, respectively. Global comparisons demonstrate consistent ADRE distributions, with relative errors below 5.8% for TOA ADRE and 4.4% for BOA ADRE. The NU-5D-LUT method significantly reduces computation time, offering a practical solution for large-scale simulations and climate model evaluations in aerosol research.

Bioaerosols are the dominant source of warm-temperature immersion-mode INPs and drive uncertainties in INP predictability.

Ice-nucleating particles (INPs) are rare atmospheric aerosols that initiate primary ice formation, but accurately simulating their concentrations and variability in large-scale climate models remains a challenge. Doing so requires both simulating major particle sources and parameterizing their ice nucleation (IN) efficiency. Validating and improving model predictions of INP concentrations requires measuring their concentrations delineated by particle type. We present a method to speciate INP concentrations into contributions from dust, sea spray aerosol (SSA), and bioaerosol. Field campaign data from Bodega Bay, California, showed that bioaerosols were the primary source of INPs between -12° and -20°C, while dust was a minor source and SSA had little impact. We found that recent parameterizations for dust and SSA accurately predicted ambient INP concentrations. However, the model did not skillfully simulate bioaerosol INPs, suggesting a need for further research to identify major factors controlling their emissions and INP efficiency for improved representation in models.

J-GAIN v1.1: a flexible tool to incorporate aerosol formation rates obtained by molecular models into large-scale models

Abstract. New-particle formation from condensable gases is a common atmospheric process that has significant but uncertain effects on aerosol particle number concentrations and aerosol–cloud–climate interactions. Assessing the formation rates of nanometer-sized particles from different vapors is an active field of research within atmospheric sciences, with new data being constantly produced by molecular modeling and experimental studies. Such data can be used in large-scale climate and air quality models through parameterizations or lookup tables. Molecular cluster dynamics modeling, ideally benchmarked against measurements when available for the given precursor vapors, provides a straightforward means to calculate high-resolution formation rate data over wide ranges of atmospheric conditions. Ideally, the incorporation of such data should be easy, efficient and flexible in the sense that same tools can be conveniently applied for different data sets in which the formation rate depends on different parameters. In this work, we present a tool to generate and interpolate lookup tables of formation rates for user-defined input parameters. The table generator primarily applies cluster dynamics modeling to calculate formation rates from an input quantum chemistry data set defined by the user, but the interpolator may also be used for tables generated by other models or data sources. The interpolation routine uses a multivariate interpolation algorithm, which is applicable to different numbers of independent parameters, and gives fast and accurate results with typical interpolation errors of up to a few percent. These routines facilitate the implementation and testing of different aerosol formation rate predictions in large-scale models, allowing the straightforward inclusion of new or updated data without the need to apply separate parameterizations or routines for different data sets that involve different chemical compounds or other parameters.

AI climate tipping‐point discovery (ACTD)

AI climate tipping‐point discovery (ACTD)

Mapping the urban heat Island at the territory scale: An unsupervised learning approach for urban planning applied to the Canton of Geneva

This study presents a fully reproducible clustering-based methodology for the assessment of the urban heat island intensity (UHII) at the territory scale, using parametric microclimate models and limited computational resources. In large-scale climate modeling, a common preliminary operation is to utilize the well-established Local Climate Zone classification to characterize the thermal response of urban areas based on morphology. With the increasing availability of urban datasets, data-driven approaches can be implemented to quantitatively derive meaningful urban features without relying on a standardized classification. The proposed methodology employs a Gaussian Mixture Model clustering algorithm to partition the urban territory into a suitable number of homogeneous microclimate zones, enabling the calculation and mapping of the UHII for each zone through the Urban Weather Generator (UWG) tool. The developed approach is applied to the Canton of Geneva, Switzerland, identifying ten microclimatic areas and analyzing the spatiotemporal variation of UHII. Results show yearly average values of UHII ranging from 1.7 °C to 2.2 °C, depending on urban morphology. The simulated values are partially validated by comparison with on-site measurements from two urban weather stations, yielding a satisfactory agreement. The methodology can support urban planning with the goal of avoid overheating through a large-scale mapping.

Understanding wind-driven melt of patchy snow cover

Abstract. The representation of snow processes in most large-scale hydrological and climate models is known to introduce considerable uncertainty into the predictions and projections of water availability. During the critical snowmelt period, the main challenge in snow modeling is that net radiation is spatially highly variable for a patchy snow cover, resulting in large horizontal differences in temperatures and heat fluxes. When a wind blows over such a system, these differences can drive advection of sensible and latent heat from the snow-free areas to the snow patches, potentially enhancing the melt rates at the leading edge and increasing the variability of subgrid melt rates. To get more insight into these processes, we examine the melt along the upwind and downwind edges of a 50 m long snow patch in the Finseelvi catchment, Norway, and try to explain the observed behavior with idealized simulations of heat fluxes and air movement over patchy snow. The melt of the snow patch was monitored from 11 June until 15 June 2019 by making use of height maps obtained through the photogrammetric structure-from-motion principle. A vertical melt of 23 ± 2.0 cm was observed at the upwind edge over the course of the field campaign, whereas the downwind edge melted only 3 ± 0.4 cm. When comparing this with meteorological measurements, we estimate the turbulent heat fluxes to be responsible for 60 % to 80 % of the upwind melt, of which a significant part is caused by the latent heat flux. The melt at the downwind edge approximately matches the melt occurring due to net radiation. To better understand the dominant processes, we represented this behavior in idealized direct numerical simulations, which are based on the measurements on a single snow patch by Harder et al. (2017) and resemble a flat, patchy snow cover with typical snow patch sizes of 15, 30, and 60 m. Using these simulations, we found that the reduction of the vertical temperature gradient over the snow patch was the main cause of the reductions in vertical sensible heat flux over distance from the leading edge, independent of the typical snow patch size. Moreover, we observed that the sensible heat fluxes at the leading edge and the decay over distance were independent of snow patch size as well, which resulted in a 15 % and 25 % reduction in average snowmelt for, respectively, a doubling and quadrupling of the typical snow patch size. These findings lay out pathways to include the effect of highly variable turbulent heat fluxes based on the typical snow patch size in large-scale hydrological and climate models to improve snowmelt modeling.

A Simple Parameterization to Enhance the Computational Time in the Three Layer Dry Deposition Model for Smooth Surfaces

Optimization of dry deposition velocity calculation has been of great interest. Every time, determining the value of the concentration boundary layer (CBL) thickness led to a waste of numerical calculation time, which appears as a huge time in large-scale climate models. The goal of this study is to optimize the numerical calculation time in the three-layer deposition model for smooth surfaces through the development of a MATLAB code that can parameterize the appropriate concentration boundary layer height (y+cbl) and internal integral calculation intervals for each particle diameter Dp (0.01–100 µm) and friction velocity u* (0.01–100 m/s). The particle concentration, as a solution to the particle flux equation, is obtained and modeled numerically by performing the left Riemann sum using MATLAB software. On the other hand, the number of subdivisions N of the Riemann sum was also parameterized for each Dp and u* in order to lessen the numerical calculation time. From a numerical point of view, the new parameterizations were tested by several computers; about 78% on the average of the computation time was saved when compared with the original algorithm. In other words, on average, about 1.2 s/calculation was gained, which is valuable in climate models simulations when millions of dry deposition calculations are needed.

Simulating wind-driven extreme sea levels: Sensitivity to wind speed and direction

Extreme sea levels and coastal flooding can cause devastating socio-economic damages and even loss of lives. Robust and timely risk mitigation requires knowledge of recurrence frequencies, but such estimates are riddled with uncertainties, especially regarding high-impact, low-probability events. In the North Sea–Baltic Sea region, considerable inter- and intra-decadal variability mean that the risks of present-day, wind-driven extreme sea levels are not fully determined. This knowledge gap is becoming more urgent for decision-makers and stakeholders since mitigation of the impacts of current and future flooding hazards, e.g. by coastal adaptation schemes, ideally requires detailed knowledge on the current and future storm-surge regime.In the present study, we investigate the wind-driven sea-level responses across the entirety of the complex Danish coastlines. We use an idealised framework to map the ranges of critical wind directions systematically by applying an advanced oceanographic model for storm surges driven by synthetic wind fields. We find that the modelled extreme water levels increase linear-to-quadratically with increased wind speed. This increase shows no signs of levelling off even for static winds of 40ms−1, although this finding is dependent on the wind stress parameterisation, which becomes uncertain for such strong wind forcing. The magnitudes of the resulting extreme water levels are highest in the Wadden Sea region in the southeastern North Sea. Still, when taking local sea-level variability into account, the most extreme incidents are undoubtedly generated in the western Baltic Sea.Hereafter, and based on the same synthetic wind fields, we investigate a complementary methodology’s potential to generalise the synthetic modelling results. For this purpose, site-specific relations between the dynamic wind forcing (i.e. wind speed and wind direction), parameters defining the local conditions and extreme sea levels are determined based on the dynamic model simulations that are generalised using a novel empirical–statistical method. The approach is tested for the North- and Baltic Seas and performs reasonably well for the highly complex coastlines around Denmark.We propose that our method could pave the way for improved operational and physically-based impact assessments based on full, large-scale regional climate model projections as a complement to select sets of coupled ocean projections.

Heat flux, water temperature and discharge from 15 northern Canadian rivers draining to Arctic Ocean and Hudson Bay

This study examined heat flux from 15 Canadian northern rivers that drain to the Arctic Ocean and the Hudson/James Bay. Based on statistical analysis of available water temperature and discharge data, we determined patterns and characteristics of discharge, water temperature, and heat flux in relation to seasonal air temperature and precipitation. We found similar seasonal cycles of discharge and water temperature across the study region, i.e. most rivers experiencing maximum discharge in June/July and highest water temperatures in July/August. The mean flows during the open water season (May to Oct.) vary from west to east along the Arctic Coast (with higher yield from the Mackenzie and Peel rivers), while river flows are higher with warmer water temperatures in western and southern Hudson Bay. The summed heat flux for the studied rivers was about 10.0 × 1012 MJ along the Arctic Coast and 2.0x1012MJ around the Hudson Bay. Among the 9 rivers flowing directly into the Arctic Ocean, the Mackenzie River with the highest flow and warmest water temperature delivered the highest heat flux, i.e. average 9.5 × 1012 MJ over the open water season during 1960–2015. These observed patterns in discharge, water temperature and heat flux were generally consistent with CHANGE model simulations for most rivers in northern Canada. The outcomes of our study provide critical knowledge of river thermal condition and heat transport to the northern seas, which will be useful for large-scale climate and ocean model development and validation, and climate/hydrology change investigations in the broader northern regions.

Cities Exacerbate Climate Warming

Urbanization is a complex land transformation not explicitly resolved within large-scale climate models. Long-term timeseries of high-resolution satellite data are essential to characterize urbanization within land surface models and to assess its contribution to surface temperature changes. The potential for additional surface warming from urbanization-induced land use change is investigated and decoupled from that due to change in climate over the continental US using a decadal timescale. We show that, aggregated over the US, the summer mean urban-induced surface temperature increased by 0.15 °C, with a warming of 0.24 °C in cities built in vegetated areas and a cooling of 0.25 °C in cities built in non-vegetated arid areas. This temperature change is comparable in magnitude to the 0.13 °C/decade global warming trend observed over the last 50 years caused by increased CO2. We also show that the effect of urban-induced change on surface temperature is felt above and beyond that of the CO2 effect. Our results suggest that climate mitigation policies must consider urbanization feedback to put a limit on the worldwide mean temperature increase.

Mesoscale wind stress-SST coupling induced feedback to the ocean in the western coast of South America

The feedback induced by mesoscale wind stress-SST coupling to the ocean in the western coast of South America was studied using the Regional Ocean Modeling System (ROMS). To represent the feedback, an empirical mesoscale wind stress perturbation model was constructed from satellite observations, and was incorporated into the ocean model. Comparing two experiments with and without the mesoscale wind stress-SST coupling, it was found that SST in the mesoscale coupling experiment was reduced in the western coast of South America, with the maximum values of 0.5 °C in the Peru Sea and 0.7 °C in the Chile Sea. A mixed layer heat budget analysis indicates that horizontal advection is the main term that explains the reduction in SST. Specifically, the feedback induced by mesoscale wind stress-SST coupling to the ocean can enhance vertical velocity in the nearshore area through the Ekman pumping, which brings subsurface cold water to the sea surface. These results indicate that the feedback due to the mesoscale wind stress-SST coupling to the ocean has the potential for reducing the warm SST bias often seen in the large-scale climate model simulations in this region.

Bayesian Analysis of Multifidelity Computer Models With Local Features and Nonnested Experimental Designs: Application to the WRF Model

ABSTRACT Motivated by a multi-fidelity Weather Research and Forecasting (WRF) climate model application where the available simulations are not generated based on hierarchically nested experimental design, we develop a new co-kriging procedure called augmented Bayesian treed co-kriging. The proposed procedure extends the scope of co-kriging in two major ways. We introduce a binary treed partition latent process in the multifidelity setting to account for nonstationary and potential discontinuities in the model outputs at different fidelity levels. Moreover, we introduce an efficient imputation mechanism which allows the practical implementation of co-kriging when the experimental design is nonhierarchically nested by enabling the specification of semiconjugate priors. Our imputation strategy allows the design of an efficient reversible jump Markov chain Monte Carlo implementation that involves collapsed blocks and direct simulation from conditional distributions. We develop the Monte Carlo recursive emulator which provides a Monte Carlo proxy for the full predictive distribution of the model output at each fidelity level, in a computationally feasible manner. The performance of our method is demonstrated on benchmark examples and used for the analysis of a large-scale climate modeling application which involves the WRF model.

The Impact of Climate Change on Wheat, Barley, and Maize Growth Indices in Near-Future and Far-Future Periods in Qazvin Plain, Iran

The current noticeable climate change that has a significant impact on the environment and agricultural systems has become a serious concern for human society. Followed by the industrial activities of the global community and increased concentrations of greenhouse gases in the atmosphere, these changes are still considered as a threat to food safety and the environment. Changes in minimum and maximum temperature, rainfall amounts, and precipitation pattern, changes in CO2 concentrations, and complex interactions between these factors and different crops are studied in the form of different probabilistic scenarios to get a glimpse of the future cultivation situation. Agriculture in Qazvin Plain is based on the water transferred from Taleqan Dam and groundwater. Since today these resources are facing issues like restriction on transfer and draw off, which is because of growth of interest in city dwelling, inappropriate water consumption patterns and over-exploitation, agriculture is sensitive to probable changes in water requirements and it is necessary to predict these changes for better management in the future. This study was carried out to investigate the effect of climate change on growth indices for three crops including wheat, barley, and maize in Qazvin Plain with different possible climate change scenarios, in the form of five large-scale climate models and three greenhouse gas emissions scenarios in 30 years’ time from 2021 to 2050 as well as from 2051 to 2080. Results showed that biomass production, grain yield, and water use efficiency in three crops of wheat, barley, and maize have increased up to 20–40% in future periods in Qazvin region climate. Growing degree-days (GDD) of these three crops have increased by 20%. Transpiration has risen by 10% in wheat and maize; however, it was stable on average in the first period and slightly decreased in the second period in barley. The transpiration coefficient has reduced by 5% for wheat and barley and has increased by 15% for maize. Results are valid as long as the initial growing conditions of these three plants are done as before, and the irrigation scheduling is performed according to the present period in spite of climate change.

Surface Phytoplankton Assemblages and Controlling Factors in the Strait of Malacca and Sunda Shelf

In the context of global warming and climate anomaly events, as a response, shifts in phytoplankton phenology were observed in the Strait of Malacca (SM) and Sunda Shelf (SS). Such interactions between phytoplankton structure and physico-chemical factors were less known in narrow straits. Therefore, the spatial distribution pattern and diversity of surface phytoplankton assemblage, local hydrology and nutrient regimes were investigated over the Strait of Malacca and Sunda Shelf (South China Sea, SCS) during 2017 and 2018 pre-monsoon season (spring). Diatoms, dinoflagellates and cyanobacteria were representatives of microphytoplankton in the survey area. Total phytoplankton abundance peaked near Singapore Strait (SGS) and diminished towards SS. From the lower ratio of diatoms to dinoflagellates (<3) in SS, we deduced lower carbon pump efficiency here. In agreement with the modeled results proposed previously, diatoms (especially centric form) tended to thrive in cold eutrophic water, while, dinoflagellates and filamentous cyanobacteria were much warm-adapted, they operated a feedback to cold La Nina event (negative Indian Ocean Dipole, IOD) in 2018 and warm El Nino event (positive IOD) in 2017, respectively. Specifically, diatom proportion increased by 30% and dinoflagellate, cyanobacteria reduced by 71%, 75% in response to shifts of climate anomaly from 2017 cruise to 2018 cruise. This interactive study between field microalgae and physico-chemiscal conditions would be helpful to launch large-scale climate model, biogeochemistry and carbon cycling in future research.

Representing surface wind stress response to mesoscale SST perturbations in western coast of South America using Tikhonov regularization method

Interaction between mesoscale perturbations of sea surface temperature (SSTmeso) and wind stress (WSmeso) has great influences on the ocean upwelling system and turbulent mixing in the atmospheric boundary layer. Using daily Quik-SCAT wind speed data and AMSR-E SST data, SSTmeso and WSmeso fields in the western coast of South America are extracted by using a locally weighted regression method (LOESS). The spatial patterns of SSTmeso and WSmeso indicate strong mesoscale SST-wind stress coupling in the region. The coupling coefficient between SSTmeso and WSmeso is about 0.009 5 N/(m2.°C) in winter and 0.008 2 N/(m2.°C) in summer. Based on mesoscale coupling relationships, the mesoscale perturbations of wind stress divergence (Div(WSmeso)) and curl (Curl (WSmeso)) can be obtained from the SST gradient perturbations, which can be further used to derive wind stress vector perturbations using the Tikhonov regularization method. The computational examples are presented in the western coast of South America and the patterns of the reconstructed WSmeso are highly consistent with SSTmeso, but the amplitude can be underestimated significantly. By matching the spatially averaged maximum standard deviations of reconstructed WSmeso magnitude and observations, a reasonable magnitude of WSmeso can be obtained when a rescaling factor of 2.2 is used. As current ocean models forced by prescribed wind cannot adequately capture the mesoscale wind stress response, the empirical wind stress perturbation model developed in this study can be used to take into account the feedback effects of the mesoscale wind stress-SST coupling in ocean modeling. Further applications are discussed for taking into account the feedback effects of the mesoscale coupling in large-scale climate models and the uncoupled ocean models.

Rate enhancement in collisions of sulfuric acid molecules due to long-range intermolecular forces

Abstract. Collisions of molecules and clusters play a key role in determining the rate of atmospheric new particle formation and growth. Traditionally the statistics of these collisions are taken from kinetic gas theory assuming spherical noninteracting particles, which may significantly underestimate the collision coefficients for most atmospherically relevant molecules. Such systematic errors in predicted new particle formation rates will also affect large-scale climate models. We studied the statistics of collisions of sulfuric acid molecules in a vacuum using atomistic molecular dynamics simulations. We found that the effective collision cross section of the H2SO4 molecule, as described by an optimized potentials for liquid simulation (OPLS). OPLS all-atom force field, is significantly larger than the hard-sphere diameter assigned to the molecule based on the liquid density of sulfuric acid. As a consequence, the actual collision coefficient is enhanced by a factor of 2.2 at 300 K compared with kinetic gas theory. This enhancement factor obtained from atomistic simulation is consistent with the discrepancy observed between experimental formation rates of clusters containing sulfuric acid and calculated formation rates using hard-sphere kinetics. We find reasonable agreement with an enhancement factor calculated from the Langevin model of capture, based on the attractive part of the atomistic intermolecular potential of mean force.

Ongoing Breakthroughs in Convective Parameterization

While the increase of computer power mobilizes a part of the atmospheric modeling community toward models with explicit convection or based on machine learning, we review the part of the literature dedicated to convective parameterization development for large-scale forecast and climate models. Many developments are underway to overcome endemic limitations of traditional convective parameterizations, either in unified or multiobject frameworks: scale-aware and stochastic approaches, new prognostic equations or representations of new components such as cold pools. Understanding their impact on the emergent properties of a model remains challenging, due to subsequent tuning of parameters and the limited understanding given by traditional metrics. Further effort still needs to be dedicated to the representation of the life cycle of convective systems, in particular their mesoscale organization and associated cloud cover. The development of more process-oriented metrics based on new observations is also needed to help quantify model improvement and better understand the mechanisms of climate change.