Global warming is a central topic in current climate change research. Since the Industrial Revolution, greenhouse gas emissions have significantly increased, particularly with a sharp rise in atmospheric CO₂ concentrations, leading to a sustained increase in global average temperatures. This change has been especially evident over the past few decades, with global temperatures rising over 1.1°C, triggering a series of climate-related issues such as accelerated glacier melting, sea-level rise, and an increase in extreme weather events. To better understand the mechanisms behind these changes, investigating Earth’s energy imbalance (EEI) has become a key focus.
Energy imbalance at the Earth's surface refers to the difference between the solar radiation absorbed by Earth and the energy radiated back into space. The increase in greenhouse gases has altered Earth’s radiation balance, trapping more energy within the Earth system. Although satellite observations have greatly improved our understanding of radiation fluxes at the top of the atmosphere (TOA), there remains considerable uncertainty about the surface radiation budget. Traditional climate models, such as CMIP6, perform well in simulating radiation budgets at the TOA but show significant biases in estimating shortwave and longwave radiation at the surface.
To address these challenges, we used Bayesian Model Averaging (BMA) in our study, a statistical method that combines multiple models with weighted averaging. BMA not only enhances estimation accuracy but also effectively reduces uncertainty, enabling us to more precisely track changes in the surface energy imbalance.
Research Methodology: Innovative Application of Bayesian Model Averaging
In our study, the BMA method was applied to constrain the outputs of CMIP6 climate models. Unlike simple ensemble averaging, BMA assigns weights to each model, reducing biases among models, especially when dependencies between different radiation components are high. We incorporated the latest surface solar radiation observations into the climate model, which allowed for a more accurate capture of EEI changes.
Our analysis results show that BMA-constrained EEI estimates align with findings from multiple international research teams for the period 2000-2014, performing better than simple ensemble averaging. Additionally, BMA reduced EEI estimation uncertainty, particularly in shortwave radiation. Since shortwave radiation contributes significantly to the uncertainty in the overall energy budget, obtaining more precise shortwave radiation observations is crucial.
Key Findings: Accelerating EEI and Its Strong Correlation with Ocean Heat Capacity
From 1961 to 2022, the increase in global surface energy imbalance has been primarily due to longwave radiation, reflecting the direct impact of greenhouse gas warming. Notably, since the mid-1990s, EEI has shown a significant upward trend. Our study indicates that BMA-estimated EEI changes are closely aligned with changes in ocean heat content (OHC), providing robust independent evidence for the accelerating trend of global warming.
Specifically, from 1961 to 1994, the average EEI value was 0.22 W/m². Excluding the influence of volcanic activity, this value increased to 0.32 W/m². However, since 1995, the average EEI value surged to 0.80 W/m², with T-test results showing that this change is statistically significant. This indicates that the intensifying trend of EEI in the late 1990s is closely associated with the rise in global temperatures during the same period.
Furthermore, our study found that the rate of EEI increase significantly accelerated after 1995. Before 1995, the EEI increase rate was 0.06 ± 0.12 W/m² per decade, whereas after 1995, this rate rose to 0.21 ± 0.09 W/m² per decade. This accelerating trend aligns closely with OHC changes. Since 1995, OHC has consistently accounted for around 90% of the EEI, indicating that the oceans have played a crucial role in absorbing excess heat.
In studying the relationship between the El Niño-Southern Oscillation (ENSO) events and global mean surface temperature, we observed some intriguing signals. After the 1997-1998 Super El Niño event, the impact of oceanic internal variability on global temperatures seems to have been overshadowed by external forcing factors, such as greenhouse gas emissions. This suggests that external forcings may be masking the effects of oceanic internal variability to some extent, reducing ENSO events’ influence on global temperatures. Although we cannot draw definitive conclusions, these observations provide clues for further exploration of OHC’s impact on the global climate system.
Conclusion and Outlook: A New Perspective on Understanding Climate Change
This study, through the innovative application of Bayesian Model Averaging, reveals the accelerating trend of Earth’s surface energy imbalance and its close relationship with ocean heat capacity. Our findings indicate that since 1995, the trend in EEI has been highly synchronized with changes in ocean heat content, providing further validation of the intensifying global warming trend.
Despite significant progress in reducing EEI estimation uncertainty, more precise observational data, especially in shortwave radiation, are still needed for future studies. This will help further improve climate models, enabling better predictions of future climate change trends.
By analyzing global warming mechanisms from the perspective of energy flow, we provide the scientific community with new tools to understand the complexities of climate change. Our study not only offers critical insights for addressing global climate change but also provides valuable references for the improvement of future climate models.
Please sign in or register for FREE
If you are a registered user on Research Communities by Springer Nature, please sign in