AHAkil Hossain

Akil Hossain

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Akil Hossain has a total of 76 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Machine Learning Eliminates Reanalysis Warm Bias and Reveals Weaker Winter Surface Cooling Over Arctic Sea Ice

    by Akil Hossain, Paul Keil, Harsh Grover, Ian M. Brooks, Christopher J. Cox, Michael R. Gallagher, Mats A. Granskog, Heather Guy, Stephen R. Hudson, P. Ola G. Persson, et al.

    The surface energy budget governs Arctic sea‐ice growth/melt, yet observations are sparse, and reanalysis data sets suffer from systematic biases. Here, we train a neural network with observational data to bias‐correct hourly ERA5 fluxes over Arctic ice‐covered regions (≥70°N; sea‐ice concentration >80%) for 1994–2024. Training data cover two full seasonal cycles and different sea‐ice regimes. The neural network reduces RMSE for net shortwave radiation by ∼40%, downward longwave radiation by ∼16% and the total surface energy budget by ∼55%, eliminating the wintertime warm bias of ∼4 K in ERA5. Wintertime surface cooling is reduced by ∼50%, yielding thermodynamic ice‐growth estimates of ∼80–120 cm, consistent with SMOS–CryoSat satellite thickness increases and in contrast to the 150–200 cm growth implied by ERA5. Our bias‐corrected data capture the observed clear/cloudy states of the winter boundary layer and can be used to study Arctic climatology, evaluate climate models and drive sea‐ice‐ocean models.

  • Open Access

    Machine Learning Eliminates Reanalysis Warm Bias and Reveals Weaker Winter Surface Cooling over Arctic Sea Ice

    by Akil Hossain, Paul Keil, Harsh Grover, Ian M. Brooks, Christopher J Cox, Michael R Gallagher, Mats A Granskog, Heather Guy, Stephen R Hudson, P Ola, et al.

    The surface energy budget governs Arctic sea-ice growth/melt, yet observations are sparse, and reanalysis datasets suffer from systematic biases. Here, we train a neural network with observational data to bias-correct hourly ERA5 fluxes over Arctic ice-covered regions (≥70°N; sea-ice concentration >80%) for 1994–2024. Training data cover two full seasonal cycles and different sea-ice regimes. The neural network reduces RMSE for net shortwave radiation by ~40%, downward longwave radiation by ~16% and the total surface energy budget by ~55%, eliminating the wintertime warm bias of ~4K in ERA5. Wintertime surface cooling is reduced by ~50%, yielding thermodynamic ice-growth estimates of ~ 80–120 cm , consistent with SMOS–CryoSat satellite thickness increases and in contrast to the 150–200 cm growth implied by ERA5. Our bias-corrected data capture the observed clear/cloudy states of the winter boundary layer and can be used to study Arctic climatology, evaluate climate models and drive sea-ice-ocean models.

  • Open Access

    Global Kilometer‐Scale Climate Storylines Using Spectral Nudging

    by Amal John, Sebastian Beyer, Marylou Athanase, Antonio Sánchez‐Benítez, Helge F. Goessling, Akil Hossain, Razvan Aguridan, Miguel Andrés‐Martínez, Aina Gaya‐ÀVila, Suvarchal Kumar Cheedela, et al.

    Effective climate adaptation benefits from detailed, actionable information on how specific extreme events are influenced by climate change. Here we present the first global, kilometer‐scale coupled modeling framework that enables counterfactual reconstructions of recent weather events, including extremes, in different climates. The system is built on the IFS–FESOM model and applies scale‐selective spectral nudging of the large‐scale atmospheric circulation to ERA5 while allowing atmospheric small‐scale dynamics and thermodynamic processes as well as the land surface, ocean, and sea ice to evolve freely across contrasting climate backgrounds. This enables the replay of recent events across different climate states, including the 1950s, present‐day conditions, and a +2 K warmer climate. We demonstrate the system's skill in reproducing the observed variability on daily and longer timescales during the period 2017–2024, particularly in the extratropics. Furthermore, we evaluate its performance using high‐quality observations, including data from the MOSAiC expedition. We further present climate change storylines for a European heatwave in July 2019 and a major European flooding event in September 2024, Storm Boris. We thereby demonstrate how the use of kilometer‐scale resolution captures local‐scale intensification with local granularity, including details missed by coarser models. This framework forms a core operational component of the EU's Destination Earth Climate Digital Twin, enabling scalable, high‐resolution, circulation‐constrained climate storylines for science and decision‐making.


  • Open Access

    Global Kilometre-Scale Climate Storylines Using Spectral Nudging

    by Amal John, Sebastian Beyer, Marylou Athanase, Antonio Sánchez Benítez, Helge Goessling, Akil Hossain, Razvan Aguridan, Miguel Andrés-Martínez, Aina Gaya-ÀVila, Suvarchal Kumar Cheedela, et al.

    Effective climate adaptation benefits from detailed, actionable information on how specific extreme events are influenced by climate change. Here we present the first global, kilometre-scale coupled modelling framework that enables counterfactual reconstructions of recent weather events, including extremes, in different climates. The system is built on the IFS–FESOM model and applies scale-selective spectral nudging of the large-scale atmospheric circulation to ERA5 while allowing atmospheric small-scale dynamics and thermodynamic processes as well as the land surface, ocean, and sea ice to evolve freely across contrasting climate backgrounds. This enables the replay of recent events across different climate states, including the 1950s, present-day conditions, and a +2 K warmer climate. We demonstrate the system's skill in reproducing the observed variability on daily and longer timescales during the period 2017--2024, particularly in the extratropics. Furthermore, we evaluate its performance using high-quality observations, including data from the MOSAiC expedition. We further present climate change storylines for a European heatwave in July 2019 and a major European flooding event in September 2024, Storm Boris. We thereby demonstrate how the use of kilometre-scale resolution captures local-scale intensification with local granularity, including details missed by coarser models. This framework forms a core operational component of the EU’s Destination Earth Climate Digital Twin, enabling scalable, high-resolution, circulation-constrained climate storylines for science and decision-making.

  • Open Access

    Machine learning eliminates near-surface warm bias in reanalysis and reveals weaker winter surface cooling over Arctic sea ice

    by Akil Hossain, Paul Keil, Harsh Grover, Ian Brooks, Christopher J. Cox, Michael R. Gallagher, Mats A. Granskog, Heather Guy, Stephen R. Hudson, P. Ola G. Persson, et al.

    The surface energy budget of the Arctic Ocean governs sea ice growth in winter and melt in summer. Understanding the surface energy budget and 2m temperature and correctly representing them in models is a key condition for understanding and projecting Arctic climate change. Direct observations of surface fluxes are scarce, and widely used reanalysis datasets suffer from systematic biases. Here, we train a neural network with observational data to bias-correct ERA5 reanalysis surface fluxes. We achieve substantial reductions in RMSE for hourly values of net shortwave radiation (~40%), downward longwave radiation (~16%) and the total surface energy budget (~55%) as well as 2m temperature (~34%). Our bias-correction eliminates the wintertime warm bias of about 4K in ERA5, reduces wintertime surface cooling by about 50% and dampens summertime surface heating. This revised surface cooling estimate is consistent with independent satellite-observed sea ice growth rates. In contrast to ERA5 fluxes, our bias-corrected data capture the observed clear and cloudy states of the Arctic winter boundary layer and the associated bimodal distribution of net longwave radiation. The bias-corrected data provide an improved baseline for climate model evaluation, climatological and case studies and forcing to drive stand-alone sea ice and ocean models.

  • Open Access

    Leveraging observations to bias-correct the Arctic surface energy budget in reanalysis through machine learning

    by Akil Hossain, Paul Keil, Harsh Grover, Felix Pithan

    The Arctic Ocean plays a critical role in global climate dynamics, yet direct observations of its surface energy budget are sparse and largely constrained to individual field campaigns. Current estimates often rely on reanalysis data, notably ERA5, which demonstrates systematic biases in boundary-layer properties and surface fluxes over Arctic sea-ice. In this study, we train an artificial neural network (ANN) to predict surface fluxes observed during MOSAiC, SHEBA, Arctic Ocean 2018 and ARTofMELT expeditions using ERA5 data as input. Data from shorter field campaigns, such as N-ICE, are used for testing our model against unseen data. Our results indicate that ERA5 Arctic surface fluxes exhibit very low correlations with observations and are characterized by large RMSE values. Our predictions demonstrate significant error reductions across key variables: sensible heat flux (~39%), 2m temperature (~39%), net shortwave radiation (~37%), downward longwave radiation (~21%) and net longwave radiation (~17%). Furthermore, we find a higher correlation (~0.58) with observations compared to ERA5 (~0.24) and approximately 50% reductions in RMSE of the hourly total surface energy budget, i.e. the sum of the individual fluxes. We produce a bias-corrected estimate of surface energy fluxes over Arctic sea-ice. We use our bias correction to revise previous estimates of the climatological surface energy budget over Arctic sea ice. We will make the trained weights available to allow for the custom derivation of bias-corrected fluxes in individual case studies and for climate model evaluation.

  • A model-based comparison of paleo-physical processes in the Bay of Bengal

    by Ashif Imam Khan, Masud- Ul- Alam, Nazrul Islam, Akil Hossain, S. M. Mustafizur Rahman

  • Open Access

    Effects of CO2 and Ocean Mixing on Miocene and Pliocene Temperature Gradients

    by Gerrit Lohmann, Gregor Knorr, Akil Hossain, Christian Stepanek

    During the Cenozoic Era hothouse climate transformed to a state that allowed establishment of extensive ice-sheets. The transformation towards an overall cooler climate encompassed periods of relatively steady change of global temperatures which were interrupted by short-term aberrations of relatively rapid cooling or warming. Various drivers have been found to contribute to this complex process of climate cooling - among these drawdown of carbon dioxide, reorganized ocean circulation related to ocean gateway evolution, varying amplitude and geographic location of deep water upwelling and formation processes, and internal feedbacks related to changes in environmental and land surface conditions in particular at high latitudes and on the continents.The fact that carbon dioxide as the most important current driver of climate change is not always proportionally linked to past changes in global temperatures underlines the importance of mechanisms beyond greenhouse gas drawdown that contributed to Cenozoic climate cooling. Several questions remain regarding mechanisms and drivers of climate evolution as reconstructed from Cenozoic proxy recorders: How can a low meridional temperature gradient be maintained at carbon dioxide concentrations that are in line with reconstructions and inference on relatively modest tropical tempatures? Which mechanisms contributed to extremely high deep sea temperatures ?Here we propose that during the Miocene and the Pliocene enhanced vertical mixing in the ocean may provide potential explanations to some of these enigmas. We employ the global general circulation model, which contributed to PlioMIP, MioMIP, and DeepMIP (e.g., Stärz et al., 2017; Stepanek et al., 2020; Hossain et al., 2020), and study the impact of variations in vertical mixing in the ocean on large-scale climate patterns, meridional temperature gradient, and deep sea ocean temperatures. We find that both carbon dioxide and enhanced vertical mixing cause increased radiative feedback by reducing effective emissivity and surface albedo. For the Miocene, enhanced oceanic heat uptake due to invigorated vertical mixing causes intense warming of the deep ocean (5-10°C) and of the Arctic (>12°C). For the Pliocene we find that the impact of radiative forcing and enhanced vertical mixing is less relevant. This hints to a dependency of carbon dioxide and mixing sensitivity to background climate and ocean dynamics.While our work is focused on climate modelling, we highlight that consideration of enhanced vertical mixing leads in our Miocene and Pliocene climate simulations to large-scale climate patterns that are in better agreement with specific aspects of proxy-based inference on past warm climates. To further corroborate our results we must compare our simulations with reconstructions of thermocline depth and seasonality - lower seasonality in reconstructions would be in line with higher heat capacity as facilitated by enhanced vertical mixing. Our ad-hoc enhanced mixing formulation for the Pliocene and Miocene (Lohmann et al., 2022) can be motivated by recent simulations with a strongly eddying ocean and an altered heat transport (Nooteboom et al, 2022). In the future, we make use of the eddy resolving model to evaluate the polar amplification of the system with respect to model resolution, gateway configuration, and background CO2.

  • Open Access

    The Impact of Different Atmospheric CO2 Concentrations on Large Scale Miocene Temperature Signatures

    by Akil Hossain, Gregor Knorr, Wilfried Jokat, Gerrit Lohmann, Katharina Hochmuth, Paul Gierz, Karsten Gohl, Christian Stepanek

    Based on inferences from proxy records the Miocene (23.03–5.33 Ma) was a time of amplified polar warmth compared to today. However, it remains a challenge to simulate a warm Miocene climate and pronounced polar warmth at reconstructed Miocene CO2 concentrations. Using a state‐of‐the‐art Earth‐System‐Model, we implement a high‐resolution paleobathymetry and simulate Miocene climate at different atmospheric CO2 concentrations. We estimate global mean surface warming of +3.1°C relative to the preindustrial at a CO2 level of 450 ppm. An increase of atmospheric CO2 from 280 to 450 ppm provides an individual warming of ∼1.4°C, which is as strong as all other Miocene forcing contributions combined. Substantial changes in surface albedo are vital to explain Miocene surface warming. Simulated surface temperatures fit well with proxy reconstructions at low‐ to mid‐latitudes. The high latitude cooling bias becomes less pronounced for higher atmospheric CO2 concentrations. At such CO2 levels simulated Miocene climate shows a reduced polar amplification, linked to a breakdown of seasonality in the Arctic Ocean. A pronounced warming in boreal fall is detected for a CO2 increase from 280 to 450 ppm, in comparison to weaker warming for CO2 changes from 450 to 720 ppm. Moreover, a pronounced warming in winter is detected for a CO2 increase from 450 to 720 ppm, in contrast to a moderate summer temperature increase, which is accompanied by a strong sea‐ice concentration decline that promotes cloud formation in summer via enhanced moisture availability. As a consequence planetary albedo increases and dampens the temperature response to CO2 forcing at a warmer Miocene background climate.


  • Open Access

    A comparison of effects of carbon dioxide and ocean mixing on Miocene and Pliocene temperature gradients

    by Gerrit Lohmann, Gregor Knorr, Akil Hossain, Christian Stepanek

    During the Cenozoic Era hothouse climate transformed to a state that allowed establishment of extensive ice-sheets. The transformation towards an overall cooler climate encompassed periods of relatively steady change of global temperatures which were interrupted by short-term aberrations of relatively rapid cooling or warming. Global climate cooling culminated in the glacial-interglacial cycles that characterize the Quaternary until today. Various drivers have been found to contribute to this complex process of climate cooling - among these drawdown of carbon dioxide, reorganized ocean circulation related to ocean gateway evolution, varying amplitude and geographic location of deep water upwelling and formation processes, and internal feedbacks related to changes in environmental and land surface conditions in particular at high latitudes and on the continents. The fact that carbon dioxide as the most important current driver of climate change is not always proportionally linked to past changes in global temperatures underlines the importance of mechanisms beyond greenhouse gas drawdown that contributed to Cenozoic climate cooling. Several questions remain regarding mechanisms and drivers of climate evolution as reconstructed from Cenozoic proxy recorders: How can a low meridional temperature gradient be maintained at carbon dioxide concentrations that are in line with reconstructions and inference on relatively modest tropical tempatures? Which mechanisms contributed to extremely high deep sea temperatures that are not commonly reproduced by simulations of Cenozoic climate? Here we propose that during the Miocene and the Pliocene enhanced vertical mixing in the ocean may provide potential explanations to some of these enigmas. We employ the global general circulation model COSMOS, which contributed to PlioMIP, MioMIP, and DeepMIP, and study the impact of variations in vertical mixing in the ocean on large-scale climate patterns, meridional temperature gradient, and deep sea ocean temperatures. We find that both carbon dioxide and enhanced vertical mixing cause increased radiative feedback by reducing effective emissivity and surface albedo. For the Miocene, enhanced oceanic heat uptake due to invigorated vertical mixing causes intense warming of the deep ocean (5-10°C) and of the Arctic (>12°C). For the Pliocene we find that the impact of radiative forcing and enhanced vertical mixing is less relevant. This hints to a dependency of carbon dioxide and mixing sensitivity to background climate and ocean dynamics. While our work is focused on climate modelling we highlight that consideration of enhanced vertical mixing leads in our Miocene and Pliocene climate simulations to large-scale climate patterns that are in better agreement with specific aspects of proxy-based inference on past warm climates. To further corroborate our results we must compare our simulations with reconstructions of thermocline depth and seasonality - lower seasonality in reconstructions would be in line with higher heat capacity as facilitated by enhanced vertical mixing.