Gentine, Pierre
Loading...
Preferred name
Gentine, Pierre
Official Name
Gentine, Pierre
ORCID
5 results
Now showing 1 - 5 of 5
- Some of the metrics are blocked by yourconsent settings
Item type:Publication, Constraining uncertainties in multi-model projections of the future climate with observations(2021-08-02); ; ; Earth system models (ESMs) are common tools to project climate change. The main focus of this thesis is the analysis of climate projections from ESMs participating in the Coupled Model Intercomparison Project (CMIP) with the aim to reduce uncertainties in climate projections with observations. In a first step, climate sensitivity is evaluated in CMIP6 models. For the effective climate sensitivity (ECS), a multi-model range of 1.8–5.6 K is found. This range is higher than in any previous CMIP ensemble before. Possible reasons for this are changes in cloud parameterizations. To reduce uncertainties in the ECS of the CMIP6 models, 11 published emergent constraints on ECS are analyzed. Emergent constraints are approaches to reduce uncertainties in climate projections by combining observations and ESM output. The application of the emergent constraints to CMIP6 data shows a decrease in the skill of the emergent relationships. This is likely related to the increased multi-model spread of ECS in CMIP6, but may in some cases also be due to spurious statistical relationships. The results support previous studies concluding that emergent constraints should be based on independently verifiable physical mechanisms. To overcome these issues of emergent constraints, an alternative approach based on machine learning (ML) is introduced. As target variable, gross primary production (GPP) is studied. In a first step, an existing emergent constraint is used to constrain the global mean GPP at the end of the 21st century in Representative Concentration Pathway (RCP) 8.5 simulations with CMIP5 ESMs to (171 ± 12) GtC yr−1. In a second step, an ML model is used to constrain gridded future absolute GPP. For this, observational data is fed into the ML algorithm that has been trained on CMIP5 data to learn relationships between present-day physically relevant diagnostics and the target variable. In a perfect model setup, the ML model shows superior performance.doctoral thesis380 463 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Toward physically consistent, accurate and stable machine learning-based convection parameterizations for ICON(2026-02-06); ; Earth system models (ESMs) are important tools to project climate change, yet continue to have persistent systematic errors due to the representation of subgrid-scale processes, most notably atmospheric convection—a key driver of large-scale circulations such as the Hadley and Walker cells, as well as weather patterns such as thunderstorms. These errors contribute considerably to uncertainties in climate projections. Traditional convection parameterizations rely on physical assumptions and empirically derived relationships that fail to capture the full complexity of convective processes. This thesis contributes to showing that machine learning (ML) offers breakthroughs, leveraging high-fidelity simulations to learn data-driven parameterizations that better represent subgrid-scale dynamics. However, translating high offline ML performance into stable, physically consistent, and transferable online implementations in ESMs has proven challenging, often due to issues related to causality, scale separation, distributional shifts, and process separation. This dissertation addresses these challenges through two complementary studies that advance the development, interpretation, and integration of ML-based convection parameterizations into the ICOsahedral Nonhydrostatic (ICON) model. The first study develops and benchmarks a suite of ML models, including deep learning and tree-based methods, trained on filtered and coarse-grained convective fluxes derived from storm-resolving ICON simulations over the tropical Atlantic. A filtering method to isolate convective contributions from other physical processes is meant to ensure that the ML models learn to represent deep convection. Offline, a U-Net architecture outperforms other models but exhibits non-causal dependencies on precipitating tracers, as revealed by explainable artificial intelligence (AI) analysis using SHapley Additive exPlanations (SHAP). Ablating these inputs yields a more physically interpretable and causally sound parameterization that demonstrates improvements in online stability, maintaining 180-day integrations in ICON while reducing biases in precipitation extremes compared to conventional schemes. However, a significant smoothing bias in the column water vapor distribution as well as biases in the mean temperature persist. Building on these insights, the second study presents a proof-of-concept for cross-model transferability and long-term integrability. A bidirectional long short-term memory model trained on the global ClimSim dataset, derived from superparameterized Energy Exascale Earth System Model-Multiscale Modeling Framework (E3SM-MMF) simulations, is transferred to the ICON-A atmosphere model. Several innovations ensure physical consistency, accuracy, and robustness: removal of radiative tendencies to isolate the convective signal, physics-informed and vertical consistency losses, confidence-guided mixing with a conventional scheme, and additive input noise during training to enhance extrapolation and stability. This hybrid AI-physics approach enables the stable multi-decadal (20-year) integration of an ML-based convection parameterization in ICON. Evaluating the resulting simulations against observations indicates improved precipitation statistics, including reduced root mean square error in the zonal mean, better spatial distribution, and improved precipitation extremes, as well as a more accurate spatial distribution of the near-surface temperature, relative to the reference ICON configuration. Furthermore, the developed scheme exhibits a physically interpretable regime behavior across column water vapor and stability metrics. However, the used training dataset has biases as well, e.g., the zonal mean precipitation does not match the observed climatology and conservation laws are not strictly enforced by the developed framework and would require a refined training dataset to do so. Moreover, the scheme is trained and evaluated at a relatively coarse horizontal resolution of ∼160 km; implementing it in the currently developed version of the hybrid ICON model, which has a horizontal resolution of ∼80 km, may require vertical interpolation and tuning. Together, these studies demonstrate that ML-based parameterizations can become stable, interpretable, and transferable components of next-generation climate models. They highlight the critical importance of physical consistency, learning causal relationships, and robust training practices in achieving reliable long-term simulations, thereby demonstrating that multi-decadal stable hybrid simulations are achievable, paving the way toward more accurate and trustworthy climate projections.doctoral thesis39 58 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Understanding and Modelling Convection with Machine Learning(2024-05-10); ; ; Machine Learning (ML) has demonstrated its potential to improve the performance of an Earth System Model (ESM), yet many challenges remain. ESMs are essential tools to project and understand climate change but have known biases. Convective processes are in general unresolved as their typical length scale is smaller than the grid size of ESMs. The effect of such subgrid processes is estimated with parameterizations, that are often attributed to be sources of biases in ESMs. A way to reduce these limitations of ESMs is to take advantage of ML or deep learning (DL) algorithms that learn actively on output from convection permitting high-resolution simulations which explicitly resolve convective processes. The resulting ML parameterizations are then coupled with an ESM and replace existing traditional subgrid parameterizations in hybrid (physics + ML) ESMs. This thesis presents novel approaches to transform DL algorithms from data science concepts towards an operational use in ESM simulations. First, a DL algorithm is developed that enables to better understand subgrid convective processes and interactions with the large-scale environment. Specifically, the latent space of a Variational Encoder Decoder (VED) reveals the meridional temperature differences between the tropics and poles together with the characteristics of subtropical and subpolar air masses along the mid-latitude storm tracks are key drivers of convective processes. Moreover, the VED separates key characteristics of shallow convective, cumulus, cirrus-like and deep convection regimes. Second, a novel DL algorithm ensemble approach is developed, that provides an improved representation of convective processes. Third, it is demonstrated that the more realistic uncertainty quantification of the ensembles capturing the chaotic nature of subgrid processes stabilizes hybrid simulations and reduces longstanding biases in a hybrid model run of an ESM. This thesis thus advances the modelling of convective processes with DL in Earth system sciences via enhanced representation and understanding of convective processes in ESMs. It provides ways to reduce limitations of state-of-the-art ML models and paves a way forward to the operational use of DL and ML in the next generation of ESMs.doctoral thesis372 142 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Data-driven cloud cover parameterizations for the ICON earth system model using deep learning and symbolic regression(2024-01-25); ; ; This thesis delves into the improvement of cloud parameterizations in climate models through machine learning trained on coarse-grained output from high-resolution simulations. Utilizing the ICOsahedral Non-hydrostatic (ICON) modeling framework, it specifically targets the enhancement of cloud cover parameterization within the ICON Earth System Model. Three types of neural networks (NNs) differing in vertical locality are developed to estimate cloud cover, with globally trained NNs even applicable to distinct regional simulations. Interpretability analysis exposes model-specific biases and local relationships with the thermodynamic environment. Despite achieving high predictive performance, NNs necessitate post-hoc interpretation tools. To tackle this issue, a combined hierarchical modeling framework incorporating symbolic regression, feature selection, and physical constraints is proposed. The resulting equations, characterized by simplicity and physical consistency, attain performance comparable to NNs while demonstrating superior transferability to other realistic datasets. Our best equation adeptly captures cloud cover distributions across various regimes, notably excelling in representing marine stratocumulus clouds by learning to utilize the vertical relative humidity gradient. This research underscores the potential of deep learning in achieving accurate cloud parameterizations and emphasizes the effective role of symbolic regression in deriving interpretable, consistent equations for cloud cover.doctoral thesis313 215 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Interpreting and improving the radiation parameterization for ICON with machine learning(2026-02-05); ; ; With progressing climate change, we need robust climate projections supporting the development of mitigation and adaptation strategies to inform policy makers. Radiation is an important part of the climate system as it regulates the energy balance of the Earth but cannot be resolved in the spectral dimension in climate models and thus needs to be parameterized. Machine Learning (ML) has been shown to be useful to improve the performance of physical parameterizations in climate models. The main focus of this thesis is to develop an ML-based radiation scheme and to investigate possible improvements of the radiation parameterization in the ICOsahedral-Nonhydrostatic (ICON) model. In the first part of this thesis, I develop an ML-based radiation emulation for the ICON model. A BiLSTM network predicts radiative heating rates and boundary fluxes at the surface and top of the atmosphere based on the current atmospheric state including boundary conditions such as incoming solar radiation. The bidirectional part of the BiLSTM scans the vertical direction of an atmospheric column, similarly to the upward and downward fluxes in physics-based radiation schemes. To analyze what the neural network has learned, I used SHapley Additive exPlanations (SHAP) to identify which input variables determine the radiative predictions of the BiLSTM. The analysis of the Shapley values reveals that the BiLSTM learned relationships related to known physical laws: a cloud reflects incoming shortwave radiation which has a strong local effect on the heating rate, but also affects all layers below leading to less heating due to less downwelling shortwave radiation below the cloud. Additionally, the layers above the cloud are affected as well, as the reflected radiation can interact with ozone. For longwave radiation, the ambient temperature plays a significant role because it determines how much longwave radiation is emitted by the gases and clouds that are locally available. This shows that the ML-based radiation emulator learns relationships that are consistent with physical processes in radiative transfer models, which is a crucial step for its operational use in an Earth System Model (ESM). The second part of this thesis focuses on the coupling of the proposed ML-based radiation emulator to the ICON model to perform hybrid simulations and test stability. The hybrid model ICON-A_ML is compared with two reference setups using low and high frequent radiation updates (ICON-A_LF, ICON-A_HF). The comparison of 10 ensemble members of each setup with a length of one year showed that ICON-A_ML has no statistically significant deviations compared to the reference simulations. A long-term simulation of 10 years demonstrates that ICON-A_ML shows no sign of accumulated biases compared to the references ICON-A_HF and ICON-A_LF. This indicates that ICON-A_ML is stable in longer simulations while also being four times faster than the original radiation parameterization. Testing with increased prescribed sea surface temperatures showed that ICON-A_ML can generalize to a +4 K warmer climate. This indicates that the ML-based radiation emulator is fit-for-purpose for its use in ICON simulations. The final study of this thesis aims to improve cloud-radiation interactions in climate models utilizing high-resolution simulations. Global Storm Resolving Models (GSRM) with a horizontal resolution of 5 km resolve cloud systems better than coarse-scale models with a horizontal resolution of around 100~km. Coarse-graining high-resolution data allows to implicitly learn subgrid-scale effects related to the horizontal and vertical distribution of clouds. I develop an approach to specifically target the improvement of the cloud radiative effects on heating rates. This approach showed 4-10 times smaller errors compared to the coarse-scale physics-based radiation scheme that was evaluated on coarse-grained data. This thesis shows that the presented ML-based radiation emulator is interpretable and that it effectively mimics physical laws. The developed ML-based radiation scheme is fast and stable when coupled to a model like ICON. Additionally, it demonstrates that high-resolution simulations can be used to improve cloud radiative effects on heating rates by implicitly learning subgrid-scale effects paving the way for reducing longstanding biases in cloud radiative effects in machine learning enhanced ESMs.doctoral thesis62 65