Abstract
Forecasts of sea surface temperature anomalies (SSTAs) provide essential information to stakeholders of marine resources in coastal ecosystems, such as the California Current Large Marine Ecosystem (CCLME), at management-relevant monthly-to-annual time scales. Diagnosing dynamical sources of predictability and the mechanisms differentiating skill among forecasts is required for verification and improvement in operational forecasting systems. Using retrospective forecasts (1982–2020) from a four-member subset of the North American Multi-Model Ensemble (NMME), we evaluate the conditional skill of SSTA forecasts in the CCLME at monthly resolution for lead times up to 10.5 months. Forecasts from ensemble members with relatively small SSTA errors at shorter lead times retain higher skill at longer lead times, with the most substantial and long-lasting increases for forecasts initialized in the fall and early spring. The “best” low-error SSTA forecasts are characterized by increased skill in the prediction of North Pacific atmospheric circulation [sea level pressure (SLP) and 200-hPa geopotential height] the month prior to the evaluation of SSTA errors in the CCLME and exhibit more realistic progressions of anomalous SLP. The Pacific meridional mode (PMM) emerges as a diagnostic of skillful North Pacific atmosphere–ocean coupling, as forecasts that correctly simulate the PMM and its associated SLP variability increase the SSTA prediction skill in the CCLME in the fall through spring. Predictable coupled ocean–atmosphere modes provide a target for enhancing predictability with early detection of the onset of a deterministic progression emerging from stochastic atmospheric variability.
Significance Statement
Global forecast systems provide near-term climate predictions that inform the management of marine resources, such as those of the California Current Large Marine Ecosystem. In this study, we probe the processes which lead forecasts to succeed or fail at predicting sea surface temperatures in the California Current at seasonal time scales among retrospective forecasts from the North American Multimodel Ensemble. We demonstrate that forecasts which best simulate sea surface temperatures at the earliest lead times sustain advantages in forecast skill and find that correctly simulating extratropical atmospheric circulation increases the predictive skill of sea surface temperatures in the northeast Pacific in the following lead times. Our results offer North Pacific atmospheric circulation as a target for forecast model improvement that would additionally enhance ocean forecasts.
Abstract
Forecasts of sea surface temperature anomalies (SSTAs) provide essential information to stakeholders of marine resources in coastal ecosystems, such as the California Current Large Marine Ecosystem (CCLME), at management-relevant monthly-to-annual time scales. Diagnosing dynamical sources of predictability and the mechanisms differentiating skill among forecasts is required for verification and improvement in operational forecasting systems. Using retrospective forecasts (1982–2020) from a four-member subset of the North American Multi-Model Ensemble (NMME), we evaluate the conditional skill of SSTA forecasts in the CCLME at monthly resolution for lead times up to 10.5 months. Forecasts from ensemble members with relatively small SSTA errors at shorter lead times retain higher skill at longer lead times, with the most substantial and long-lasting increases for forecasts initialized in the fall and early spring. The “best” low-error SSTA forecasts are characterized by increased skill in the prediction of North Pacific atmospheric circulation [sea level pressure (SLP) and 200-hPa geopotential height] the month prior to the evaluation of SSTA errors in the CCLME and exhibit more realistic progressions of anomalous SLP. The Pacific meridional mode (PMM) emerges as a diagnostic of skillful North Pacific atmosphere–ocean coupling, as forecasts that correctly simulate the PMM and its associated SLP variability increase the SSTA prediction skill in the CCLME in the fall through spring. Predictable coupled ocean–atmosphere modes provide a target for enhancing predictability with early detection of the onset of a deterministic progression emerging from stochastic atmospheric variability.
Significance Statement
Global forecast systems provide near-term climate predictions that inform the management of marine resources, such as those of the California Current Large Marine Ecosystem. In this study, we probe the processes which lead forecasts to succeed or fail at predicting sea surface temperatures in the California Current at seasonal time scales among retrospective forecasts from the North American Multimodel Ensemble. We demonstrate that forecasts which best simulate sea surface temperatures at the earliest lead times sustain advantages in forecast skill and find that correctly simulating extratropical atmospheric circulation increases the predictive skill of sea surface temperatures in the northeast Pacific in the following lead times. Our results offer North Pacific atmospheric circulation as a target for forecast model improvement that would additionally enhance ocean forecasts.
Abstract
Based on the observations and the phase 6 of Coupled Model Intercomparison Project (CMIP6) multimodel simulations, we conducted a detection and attribution analysis for the observed changes in intensity and frequency indices of extreme precipitation during 1961–2014 over the whole of China and within distinct climate regions across the country. A space–time analysis is simultaneously applied in detection so that spatial structure on the signals is considered. Results show that the CMIP6 models can simulate the observed general increases of extreme precipitation indices during the historical period except for the drying trends from southwestern to northeastern China. The anthropogenic (ANT) signal is detectable and attributable to the observed increase of extreme precipitation over China, with human-induced greenhouse gas (GHG) increases being the dominant contributor. Additionally, we also detected the ANT and GHG signals in China’s temperate continental, subtropical–tropical monsoon, and plateau mountain climate zones, demonstrating the role of human activity in historical extreme precipitation changes on much smaller spatial scales.
Significance Statement
The observed intensification of extreme precipitation globally has been attributed to human influences. Here, we demonstrate that anthropogenic forcing has discernably intensified extreme precipitation over the period 1961–2014, over China and in three of its four climate zones, with human-induced greenhouse gas increases being the dominant contributor. Our results strengthen the body of evidence that greenhouse gas increases are intensifying extreme precipitation by quantifying their role in observed changes at smaller regional scales than previously reported.
Abstract
Based on the observations and the phase 6 of Coupled Model Intercomparison Project (CMIP6) multimodel simulations, we conducted a detection and attribution analysis for the observed changes in intensity and frequency indices of extreme precipitation during 1961–2014 over the whole of China and within distinct climate regions across the country. A space–time analysis is simultaneously applied in detection so that spatial structure on the signals is considered. Results show that the CMIP6 models can simulate the observed general increases of extreme precipitation indices during the historical period except for the drying trends from southwestern to northeastern China. The anthropogenic (ANT) signal is detectable and attributable to the observed increase of extreme precipitation over China, with human-induced greenhouse gas (GHG) increases being the dominant contributor. Additionally, we also detected the ANT and GHG signals in China’s temperate continental, subtropical–tropical monsoon, and plateau mountain climate zones, demonstrating the role of human activity in historical extreme precipitation changes on much smaller spatial scales.
Significance Statement
The observed intensification of extreme precipitation globally has been attributed to human influences. Here, we demonstrate that anthropogenic forcing has discernably intensified extreme precipitation over the period 1961–2014, over China and in three of its four climate zones, with human-induced greenhouse gas increases being the dominant contributor. Our results strengthen the body of evidence that greenhouse gas increases are intensifying extreme precipitation by quantifying their role in observed changes at smaller regional scales than previously reported.
Abstract
In the Southern Hemisphere, Earth system models project an intensification of winter storm tracks by the end of the twenty-first century. Previous studies using idealized models showed that storm track intensity saturates with increasing temperatures, suggesting that the intensification of the winter storm tracks might not continue further with increasing greenhouse gases. Here, we examine the response of midlatitude winter storm tracks in the Southern Hemisphere to increasing CO2 from two to eight times preindustrial concentrations in more realistic Earth system models. We find that at high CO2 levels (beyond 4×CO2), winter storm tracks no longer exhibit an intensification across the extratropics. Instead, they shift poleward, weakening the storm tracks at lower midlatitudes and strengthening at higher midlatitudes. By analyzing the eddy kinetic energy (EKE) budget, the nonlinear storm-track response to an increase in CO2 levels in the lower midlatitudes is found to stem from a scale-dependent conversion of eddy available potential energy to EKE. Specifically, in the lower midlatitudes, this energy conversion acts to oppositely change the EKE of long and short scales at low CO2 levels, but at high CO2 levels, it mostly reduces the EKE of shorter scales, resulting in a poleward shift of the storms. Furthermore, we identify a “tug of war” between the upper and lower temperature changes as the primary driver of the nonlinear-scale-dependent EKE response in the lower midlatitudes. Our results suggest that in the highest emission scenarios beyond the twenty-first century, the storm tracks’ response may differ in magnitude and latitudinal distribution from projected changes by 2100.
Significance Statement
The Southern Hemisphere winter storm track is projected to intensify by the end of the century, with the most significant intensification occurring in the higher midlatitudes. However, we show that the intensification is not a linear function of the radiative forcing associated with increasing CO2 levels. In fact, our study shows a poleward shift at very high CO2 levels, with the storm track moving southward. This suggests that the Southern Hemisphere winter storm track may require time-sensitive adaptation strategies, as the impacts of global warming on the storm track may not be a linear function of CO2 concentration in the atmosphere.
Abstract
In the Southern Hemisphere, Earth system models project an intensification of winter storm tracks by the end of the twenty-first century. Previous studies using idealized models showed that storm track intensity saturates with increasing temperatures, suggesting that the intensification of the winter storm tracks might not continue further with increasing greenhouse gases. Here, we examine the response of midlatitude winter storm tracks in the Southern Hemisphere to increasing CO2 from two to eight times preindustrial concentrations in more realistic Earth system models. We find that at high CO2 levels (beyond 4×CO2), winter storm tracks no longer exhibit an intensification across the extratropics. Instead, they shift poleward, weakening the storm tracks at lower midlatitudes and strengthening at higher midlatitudes. By analyzing the eddy kinetic energy (EKE) budget, the nonlinear storm-track response to an increase in CO2 levels in the lower midlatitudes is found to stem from a scale-dependent conversion of eddy available potential energy to EKE. Specifically, in the lower midlatitudes, this energy conversion acts to oppositely change the EKE of long and short scales at low CO2 levels, but at high CO2 levels, it mostly reduces the EKE of shorter scales, resulting in a poleward shift of the storms. Furthermore, we identify a “tug of war” between the upper and lower temperature changes as the primary driver of the nonlinear-scale-dependent EKE response in the lower midlatitudes. Our results suggest that in the highest emission scenarios beyond the twenty-first century, the storm tracks’ response may differ in magnitude and latitudinal distribution from projected changes by 2100.
Significance Statement
The Southern Hemisphere winter storm track is projected to intensify by the end of the century, with the most significant intensification occurring in the higher midlatitudes. However, we show that the intensification is not a linear function of the radiative forcing associated with increasing CO2 levels. In fact, our study shows a poleward shift at very high CO2 levels, with the storm track moving southward. This suggests that the Southern Hemisphere winter storm track may require time-sensitive adaptation strategies, as the impacts of global warming on the storm track may not be a linear function of CO2 concentration in the atmosphere.
Abstract
An Ensemble Tangent Linear Model (ETLM) is applied to a cloud physics scheme used in the Navy Global Environmental Model (NAVGEM). The ensemble is created using 3-hour forecasts from the Ensemble Transform method used in the NAVGEM data assimilation system. The model states are saved before and after applying the cloud physics parameterization (which includes condensation/evaporation of cloud ice and cloud liquid water and stratiform precipitation), and these states are used to construct linearized model tendencies for temperature, specific humidity, cloud liquid water, and cloud ice water. We examine separately the application of the ETLM to cloud physics components that are explicitly local versus non-local. For the local components, an ETLM is built using a single grid point. ETLMs from 50 to 1000 members are tested, and skillful forecasts can be obtained for both local and non-local physics even with a moderate sized ensemble (e.g., 100 members). At 1000 members, the globally-averaged forecast error reductions (relative to persistence errors) are ∼40% for temperature, water vapor, and cloud liquid water and ∼30% for cloud ice. When initial perturbations are reduced by a factor of 0.1, the error reductions increase to ∼65% for all variables. For physics with non-local components (stratiform precipitation) the covariances that comprise the ETLM are localized with a Schur product matrix using a Gaussian localization shape with tunable length. The optimal lengths increased with ensemble size from ∼2-3 km for 50 members to ∼10 km for 1000 members. ETLMs for “all cloud physics” are also constructed and evaluated.
Abstract
An Ensemble Tangent Linear Model (ETLM) is applied to a cloud physics scheme used in the Navy Global Environmental Model (NAVGEM). The ensemble is created using 3-hour forecasts from the Ensemble Transform method used in the NAVGEM data assimilation system. The model states are saved before and after applying the cloud physics parameterization (which includes condensation/evaporation of cloud ice and cloud liquid water and stratiform precipitation), and these states are used to construct linearized model tendencies for temperature, specific humidity, cloud liquid water, and cloud ice water. We examine separately the application of the ETLM to cloud physics components that are explicitly local versus non-local. For the local components, an ETLM is built using a single grid point. ETLMs from 50 to 1000 members are tested, and skillful forecasts can be obtained for both local and non-local physics even with a moderate sized ensemble (e.g., 100 members). At 1000 members, the globally-averaged forecast error reductions (relative to persistence errors) are ∼40% for temperature, water vapor, and cloud liquid water and ∼30% for cloud ice. When initial perturbations are reduced by a factor of 0.1, the error reductions increase to ∼65% for all variables. For physics with non-local components (stratiform precipitation) the covariances that comprise the ETLM are localized with a Schur product matrix using a Gaussian localization shape with tunable length. The optimal lengths increased with ensemble size from ∼2-3 km for 50 members to ∼10 km for 1000 members. ETLMs for “all cloud physics” are also constructed and evaluated.
Abstract
We analyzed 14 days of observations from sonic anemometry and high-resolution fiber optic distributed sensing collected in the stable polar boundary layer (SBL). The study sought to evaluate if and under which conditions the sensible heat flux is related to the temperature gradient. Machine learning methods were employed to identify drivers of and model heat fluxes. We found the recently proposed coupling metric Ω defined as the ratio of the buoyancy length scale and measurement height to delineate physically meaningful transport regimes. The regime transition marks the point where static stability in addition to the vertical turbulence strength control the heat transport, which is rather gradual than abrupt. The maximum downward heat flux is reached when one third of turbulent eddies exceed the opposing buoyancy forces in the SBL. We found evidence that even for large Ω a substantial fraction of the turbulent transport is non-equilibrium. The non-dimensional temperature gradient is better explained by variations in Ω than ζ = zL −1 from Monin-Obukhov Similarity theory. Its continuous organization with Ω across stabilities suggest that the vertical heat transport always remains coupled to the surface, but its efficiency and the resulting flux vary. 43% of the total enthalpy is exchanged during conditions of limited transport efficiency in the very SBL despite the small flux magnitude of ≤ 7 W m−2, which underlines the importance of quantifying the weak surface exchange for polar regions. When predicting sensible heat fluxes using mean quantities from weather stations, the net longwave radiative forcing and the horizontal wind speed are the most important predictors representing stratification and bulk shear.
Abstract
We analyzed 14 days of observations from sonic anemometry and high-resolution fiber optic distributed sensing collected in the stable polar boundary layer (SBL). The study sought to evaluate if and under which conditions the sensible heat flux is related to the temperature gradient. Machine learning methods were employed to identify drivers of and model heat fluxes. We found the recently proposed coupling metric Ω defined as the ratio of the buoyancy length scale and measurement height to delineate physically meaningful transport regimes. The regime transition marks the point where static stability in addition to the vertical turbulence strength control the heat transport, which is rather gradual than abrupt. The maximum downward heat flux is reached when one third of turbulent eddies exceed the opposing buoyancy forces in the SBL. We found evidence that even for large Ω a substantial fraction of the turbulent transport is non-equilibrium. The non-dimensional temperature gradient is better explained by variations in Ω than ζ = zL −1 from Monin-Obukhov Similarity theory. Its continuous organization with Ω across stabilities suggest that the vertical heat transport always remains coupled to the surface, but its efficiency and the resulting flux vary. 43% of the total enthalpy is exchanged during conditions of limited transport efficiency in the very SBL despite the small flux magnitude of ≤ 7 W m−2, which underlines the importance of quantifying the weak surface exchange for polar regions. When predicting sensible heat fluxes using mean quantities from weather stations, the net longwave radiative forcing and the horizontal wind speed are the most important predictors representing stratification and bulk shear.
Abstract
Multi-model ensemble forecasts have gained widespread use over the past decade. A yet unresolved issue is whether forecast skill benefits from the use of prior skill from each model in providing a weighted combination. Here we use the available seasonal ensemble forecasts of six models from the North American Multi-Model Ensemble (NMME) to study various aspects of prior skill-based weighting schemes and explore ways to merge multi-model forecasts. First, we post-process each NMME model through quantile mapping and a simple spread error adjustment. Then, using an equal weighted combination as the baseline forecast, we test merging the models together through skill-based weights by varying the prior skill metric and varying how the metrics are aggregated across the different subbasins and time of year. Results confirm prior work that the combined forecasts do outperform individual models. When evaluating prior skill, equal weighting generally performed as well or slightly better than all weighting schemes tried. The skill of the weighting scheme was not found to be strongly dependent on prior metric but did improve when aggregating all forecasted months and subbasins together to provide one overall weight to each model. Also, we found that including an offset to the prior metric that nudged the weights closer to equal weighting improves skill especially at longer leads where individual model skill is low. Results also show that the weighting schemes performed better than regression-based techniques including multiple linear regression and random forest.
Abstract
Multi-model ensemble forecasts have gained widespread use over the past decade. A yet unresolved issue is whether forecast skill benefits from the use of prior skill from each model in providing a weighted combination. Here we use the available seasonal ensemble forecasts of six models from the North American Multi-Model Ensemble (NMME) to study various aspects of prior skill-based weighting schemes and explore ways to merge multi-model forecasts. First, we post-process each NMME model through quantile mapping and a simple spread error adjustment. Then, using an equal weighted combination as the baseline forecast, we test merging the models together through skill-based weights by varying the prior skill metric and varying how the metrics are aggregated across the different subbasins and time of year. Results confirm prior work that the combined forecasts do outperform individual models. When evaluating prior skill, equal weighting generally performed as well or slightly better than all weighting schemes tried. The skill of the weighting scheme was not found to be strongly dependent on prior metric but did improve when aggregating all forecasted months and subbasins together to provide one overall weight to each model. Also, we found that including an offset to the prior metric that nudged the weights closer to equal weighting improves skill especially at longer leads where individual model skill is low. Results also show that the weighting schemes performed better than regression-based techniques including multiple linear regression and random forest.
Abstract
Improving Land Surface Temperature (LST) modeling is vital for mitigating climate change effects on various ecosystems and marine habitats such as on important sea turtle habitats. Over the past decade, extreme temperatures have likely significantly affected nesting sea turtle habitats in the Arabian Gulf, with predominantly female hatchlings creating an imbalance in the sex ratio. Such shifts have profound implications for these habitats' long-term survival and conservation management. This study leverages statistical machine learning models to measure ongoing temporal variations in LST. We break down the LST time series into trend, seasonal, and noise components using classical decomposition methods like X11, SEATS, and the Seasonal and Trend decomposition using Loess (STL) approach. The long-term trends in LST data are driven by climate change rather than seasonal fluctuations. We employed Neural Network Auto Regression (NNAR), BaggedETS, Exponential Smoothing models, and STL method to project future LST values. We also explored advanced forecasting models like Dynamic Harmonic Regression, TBATS, and SARIMA for comparative performance analysis. Extended warm periods were identified for Abu Ali Island between 2017 and 2018 through several decomposition methods, likely linked to the 2015-2016 El Niño event. We also conducted a Marine Heat Wave (MHW) analysis from 2010-2020, establishing a pronounced impact of the 2015-2016 El Niño on the Arabian Gulf. In nesting beach environments with high LST, marine heatwaves could have a significant impact on sea turtle populations without human intervention such as artificially cooling the nest temperature. SARIMA model showed higher forecasting precision for in-situ weather data while NNAR model demonstrated superior performance with remotely sensed data.
Abstract
Improving Land Surface Temperature (LST) modeling is vital for mitigating climate change effects on various ecosystems and marine habitats such as on important sea turtle habitats. Over the past decade, extreme temperatures have likely significantly affected nesting sea turtle habitats in the Arabian Gulf, with predominantly female hatchlings creating an imbalance in the sex ratio. Such shifts have profound implications for these habitats' long-term survival and conservation management. This study leverages statistical machine learning models to measure ongoing temporal variations in LST. We break down the LST time series into trend, seasonal, and noise components using classical decomposition methods like X11, SEATS, and the Seasonal and Trend decomposition using Loess (STL) approach. The long-term trends in LST data are driven by climate change rather than seasonal fluctuations. We employed Neural Network Auto Regression (NNAR), BaggedETS, Exponential Smoothing models, and STL method to project future LST values. We also explored advanced forecasting models like Dynamic Harmonic Regression, TBATS, and SARIMA for comparative performance analysis. Extended warm periods were identified for Abu Ali Island between 2017 and 2018 through several decomposition methods, likely linked to the 2015-2016 El Niño event. We also conducted a Marine Heat Wave (MHW) analysis from 2010-2020, establishing a pronounced impact of the 2015-2016 El Niño on the Arabian Gulf. In nesting beach environments with high LST, marine heatwaves could have a significant impact on sea turtle populations without human intervention such as artificially cooling the nest temperature. SARIMA model showed higher forecasting precision for in-situ weather data while NNAR model demonstrated superior performance with remotely sensed data.
Abstract
The paper aims at investigating the effectiveness of estimating vertical profiles of air temperature and PM10 concentrations in Alpine valleys through ground stations positioned at different altitudes on one valley sidewall (i.e. pseudo-vertical profiles). Two case studies in the Italian Alps are investigated: Chiese Valley in Trentino province and Camonica Valley in Lombardy region. Vertical profiles of temperature and PM10 concentrations were derived from airborne measurements at the center of the two valleys by means of low-cost sensors installed on a drone during summer 2019 and a tethered balloon during winter 2020. At the same time, five stations, equipped with the same kind of low-cost sensors, simultaneously monitored the same variables on one mountain slope. Comparisons between pseudo-profiles and airborne soundings revealed that ground stations well approximated temperature and PM10 soundings during the night and early morning, while temperatures along the slopes were higher than in the center of the valley during daytime, due to solar radiative heating, with larger differences in summer than in winter. On the contrary, some episodes with PM10 concentrations slightly higher in the valley center than on the slope were recorded, due to transport events and upslope winds. Nonetheless, the pseudo-profiles based on slope ground measurements faithfully reproduced the vertical gradients of both air temperature and PM10 if compared to those assessed from the soundings performed at the center of the two valleys. Results show that pseudo-vertical profiles can be a reliable and inexpensive method for continuous monitoring of vertical air temperature and PM10 distribution in mountain valleys.
Abstract
The paper aims at investigating the effectiveness of estimating vertical profiles of air temperature and PM10 concentrations in Alpine valleys through ground stations positioned at different altitudes on one valley sidewall (i.e. pseudo-vertical profiles). Two case studies in the Italian Alps are investigated: Chiese Valley in Trentino province and Camonica Valley in Lombardy region. Vertical profiles of temperature and PM10 concentrations were derived from airborne measurements at the center of the two valleys by means of low-cost sensors installed on a drone during summer 2019 and a tethered balloon during winter 2020. At the same time, five stations, equipped with the same kind of low-cost sensors, simultaneously monitored the same variables on one mountain slope. Comparisons between pseudo-profiles and airborne soundings revealed that ground stations well approximated temperature and PM10 soundings during the night and early morning, while temperatures along the slopes were higher than in the center of the valley during daytime, due to solar radiative heating, with larger differences in summer than in winter. On the contrary, some episodes with PM10 concentrations slightly higher in the valley center than on the slope were recorded, due to transport events and upslope winds. Nonetheless, the pseudo-profiles based on slope ground measurements faithfully reproduced the vertical gradients of both air temperature and PM10 if compared to those assessed from the soundings performed at the center of the two valleys. Results show that pseudo-vertical profiles can be a reliable and inexpensive method for continuous monitoring of vertical air temperature and PM10 distribution in mountain valleys.
Abstract
The Atlantic Meridional Overturning Circulation (AMOC) is an important component of the global climate, known to be a tipping element, as it could collapse under global warming. The main objective of this study is to compute the probability that the AMOC collapses within a specified time window, using a rare-event algorithm called Trajectory-Adaptive Multilevel Splitting (TAMS). However, the efficiency and accuracy of TAMS depend on the choice of the score function. Although the definition of the optimal score function, called “committor function” is known, it is impossible in general to compute it a priori. Here, we combine TAMS with a Next-Generation Reservoir Computing technique that estimates the committor function from the data generated by the rare-event algorithm. We test this technique in a stochastic box model of the AMOC for which two types of transition exist, the so-called F(ast)-transitions and S(low)-transitions. Results for the F-transtions compare favorably with those in the literature where a physically-informed score function was used. We show that coupling a rare-event algorithm with machine learning allows for a correct estimation of transition probabilities, transition times, and even transition paths for a wide range of model parameters. We then extend these results to the more difficult problem of S-transitions in the same model. In both cases of F-transitions and S-transitions, we also show how the Next-Generation Reservoir Computing technique can be interpreted to retrieve an analytical estimate of the committor function.
Abstract
The Atlantic Meridional Overturning Circulation (AMOC) is an important component of the global climate, known to be a tipping element, as it could collapse under global warming. The main objective of this study is to compute the probability that the AMOC collapses within a specified time window, using a rare-event algorithm called Trajectory-Adaptive Multilevel Splitting (TAMS). However, the efficiency and accuracy of TAMS depend on the choice of the score function. Although the definition of the optimal score function, called “committor function” is known, it is impossible in general to compute it a priori. Here, we combine TAMS with a Next-Generation Reservoir Computing technique that estimates the committor function from the data generated by the rare-event algorithm. We test this technique in a stochastic box model of the AMOC for which two types of transition exist, the so-called F(ast)-transitions and S(low)-transitions. Results for the F-transtions compare favorably with those in the literature where a physically-informed score function was used. We show that coupling a rare-event algorithm with machine learning allows for a correct estimation of transition probabilities, transition times, and even transition paths for a wide range of model parameters. We then extend these results to the more difficult problem of S-transitions in the same model. In both cases of F-transitions and S-transitions, we also show how the Next-Generation Reservoir Computing technique can be interpreted to retrieve an analytical estimate of the committor function.
