What is North Atlantic Deep Water formation and why does it matter for the climate?

North Atlantic Deep Water formation is a process where surface water becomes dense enough to sink to great depths. This sinking is a crucial part of ocean ventilation, bringing oxygen to the deep ocean and influencing the global ocean circulation. Its significance lies in its role in carbon and heat uptake from the atmosphere, which affects the rate of sea level rise. How well climate models represent this formation directly impacts the accuracy of predictions regarding Arctic oceanic warming and its downstream consequences.

What is the Atlantic Meridional Overturning Circulation (AMOC) and why is it important for understanding climate change?

The Atlantic Meridional Overturning Circulation, or AMOC, is a system of ocean currents that transports heat from the tropics towards the Arctic. The AMOC is important because it influences sea ice melt and the stability of glaciers, especially in Greenland. If climate models inaccurately simulate North Atlantic Deep Water formation, which drives the AMOC, projections of future climate, particularly in the Arctic region, may not be reliable.

What is deep convection, and why is it important for climate models to get it right?

Deep convection refers to the process where surface water sinks due to increased density. It's important in the context of climate because it facilitates the exchange of heat, carbon, and other properties between the surface and deep ocean. Climate models often struggle to accurately simulate this process, leading to potential biases in predicting future climate scenarios. Accurately simulating deep convection is essential for understanding and predicting changes in ocean temperature, salinity, and circulation patterns.

What is CMIP5, and what does it tell us about the ability of climate models to simulate deep water formation?

CMIP5, or the Climate Model Intercomparison Project phase 5, is a collaborative effort involving numerous climate modeling groups worldwide. It provides a framework for comparing and evaluating climate models. The models in CMIP5 have shown biases in representing North Atlantic Deep Water formation. Addressing these biases is important to improve the reliability of climate projections and better prepare for the consequences of climate change.

How does the simulation of sea ice extent impact the accuracy of deep convection in climate models?

Sea ice extent is the area of ocean covered by sea ice. Its simulation is very important because it influences deep convection. Models that accurately represent sea ice extent, especially those using the CICE sea ice model, tend to simulate deep convection more realistically. The accuracy of sea ice representation in climate models is crucial for predicting changes in Arctic oceanic warming and its impact on global climate.

Surreal illustration of North Atlantic ocean currents and Arctic sea ice depicting climate change.

Navigating Climate Change: Unveiling the Secrets of Deep Ocean Currents and Their Impact

Avery Sinclair in Climate & Environment December 2025 • 4 min read.

"A deep dive into how North Atlantic deep water formation and AMOC are represented in CMIP5 models, revealing critical insights for more accurate climate predictions."

Global climate models are essential for understanding and predicting climate change, yet they often struggle with biases that affect their accuracy. One key area of concern is the representation of deep water formation, a process vital for ocean circulation, carbon and heat uptake, and ultimately, sea level rise.

Deep water formation occurs in regions like Antarctica and the North Atlantic, playing a crucial role in ocean ventilation and the global ocean circulation. In the North Atlantic, this process is closely linked to the Atlantic Meridional Overturning Circulation (AMOC), which transports heat towards the Arctic. This heat influences sea ice melt and the stability of Greenland's glaciers, making the North Atlantic a critical area for evaluating the performance of climate models.

This article explores a detailed comparison of deep water formation in 23 state-of-the-art global climate models from the Climate Model Intercomparison Project phase 5 (CMIP5). By assessing the biases in their representation of deep convection, examining the causes, and estimating the consequences for AMOC and Arctic heat export, we aim to highlight the strengths and weaknesses of current-generation climate models. Understanding these dynamics is essential for realistically forecasting Arctic oceanic warming and its far-reaching impacts.

Deep Convection in Climate Models: Are They Getting It Right?

Surreal illustration of North Atlantic ocean currents and Arctic sea ice depicting climate change.

The study reveals that most climate models struggle to accurately simulate deep convection, a process where surface water sinks due to increased density. Most models tend to simulate convection that is too deep, over too large an area, too frequent, and too far south. This skewed representation highlights the challenges in capturing the complex dynamics of real-world deep water formation.

Interestingly, the research found that deep convection often occurs at the sea ice edge and is most realistically captured in models that accurately simulate sea ice extent. Models using the CICE sea ice model tended to perform better in this regard, suggesting that accurate sea ice representation is crucial for simulating deep convection.

Overestimated Convection: Most models simulate deep convection that extends too far, both in depth and geographical area, compared to observational data.
Inaccurate Location: Many models misplace the location of deep convection, particularly in the subpolar gyre, where convection often occurs too far south.
Frequency Issues: Models tend to simulate deep convection every year, unlike the real ocean, where it occurs intermittently.

Further analysis revealed that about half of the models simulate convection in response to local cooling or salinification of surface waters, while only a third capture the dynamic relationship between freshwater input from the Arctic and deep convection. Models with the most intense deep convection tend to have the warmest deep waters, a result of heat redistribution through the water column.

Why Understanding Model Biases Matters for the Future

Ultimately, understanding the dynamical drivers of deep convection and AMOC in climate models is essential for accurately forecasting Arctic oceanic warming. By addressing the biases and limitations in current models, scientists can improve the reliability of climate projections and better prepare for the far-reaching consequences of Arctic change on global ocean circulation, the cryosphere, and marine life. Further model intercomparison efforts and dedicated studies are needed to refine these representations and enhance our predictive capabilities.