Subtropical Low Cloud Response to a Warmer Climate in a Superparameterized Climate Model. Part II: Column Modeling with a Cloud Resolving Model

Peter N. Blossey, Christopher S. Bretherton, M. C. WyantJournal of Advances in Modeling Earth Systems 2009 · DOI

TL;DR

This study uses a cloud-resolving model (CRM) in a column-modeling framework to investigate how subtropical low clouds respond to a 2K warming. They found that the resolution of the CRM significantly impacts the simulated cloud response, suggesting that current climate models might overestimate negative cloud feedbacks due to insufficient resolution of trade cumulus clouds.

What Problem Does This Paper Address?

Climate models struggle to accurately represent clouds, particularly low clouds in the subtropics, which play a crucial role in regulating Earth's temperature. These clouds are thought to provide a negative feedback (cooling effect) in a warming climate, but the magnitude of this feedback is uncertain. Superparameterized climate models (SP-CAMs) embed cloud-resolving models (CRMs) within each grid cell to better represent clouds. However, the resolution of these embedded CRMs is a computational compromise. This paper addresses the question of whether the resolution of the CRM within SP-CAM is sufficient to accurately capture the response of subtropical low clouds to climate warming, and if not, how this under-resolution might be exaggerating the simulated negative cloud feedback.

Methods

The researchers employed a column-modeling approach using the System for Atmospheric Modeling (SAM) CRM. This method isolates a single vertical column of the atmosphere and forces it with large-scale atmospheric conditions derived from a superparameterized climate model (SP-CAM). Key aspects of their methodology include:

  • SP-CAM Composites: They utilized monthly-mean output from SP-CAM simulations for both a control climate and a +2K warmer climate. These outputs were sorted into different cloud regimes based on lower-tropospheric stability (LTS).
  • Weak Temperature Gradient (WTG) Algorithm: A novel algorithm, based on an idealized form of gravity wave adjustment, was used to adjust vertical motion within the CRM column. This ensured that the simulated virtual temperature profile remained consistent with the corresponding SP-CAM composite profile, mimicking the large-scale tendency towards a moist-adiabatic profile in the tropics.
  • Humidity Nudging: Humidity above the boundary layer was slowly relaxed towards the SP-CAM composite profile to prevent unrealistic drifts.
  • Resolution Sensitivity: The core of the study involved running the CRM at various horizontal resolutions (from 4 km down to 50 m) to assess the impact of resolution on the simulated cloud response.

Datasets

The primary dataset used in this study is output from the SP-CAM (Superparameterized Climate Model) simulations. Specifically, they used monthly-mean data from control and +2K sea surface temperature (SST) experiments, sorted into different cloud regimes based on lower-tropospheric stability (LTS). They also referenced the RICO wind profile for initializing wind conditions.

Key Insights and Ideas

  • Resolution Matters (A Lot): The most striking finding is how dramatically the simulated low cloud response to warming changes with CRM resolution. Coarser resolutions (like the 4 km used in SP-CAM) show a significant increase in low cloud cover and associated cooling (negative feedback), while finer resolutions (LES scales) show a much smaller increase, or even a decrease in some cases.
  • Overestimation of Negative Feedback: The study suggests that the negative cloud feedback simulated by SP-CAM might be exaggerated because the CRM's resolution is too coarse to properly represent the physics of trade cumulus boundary layers.
  • Trade Inversion Dominance at Fine Scales: At very fine resolutions, the simulated cloud increase in a warmer climate appears to be localized near the trade inversion, suggesting that a stronger inversion might trap more condensate.
  • Column Analogue Utility: The column modeling approach proves to be a valuable tool for isolating the impact of CRM resolution and for understanding the physical mechanisms driving cloud changes, bridging the gap between GCMs and full CRM simulations.

Key Results

  • CRM4km vs. SP-CAM: When the CRM was run at a resolution comparable to SP-CAM (4 km), it reproduced a similar +2K increase in low cloud fraction and shortwave cloud forcing (SWCF) as observed in SP-CAM.
  • Fine Resolution (LES): At finer resolutions (e.g., 100 m), the CRM-simulated low cloud fraction and its increase in a warmer climate were significantly smaller. The +2K change in SWCF was also substantially reduced compared to the 4 km CRM and SP-CAM.
  • Trade Cumulus Biases: Coarse CRM resolution (4 km) leads to an overestimation of cloud fraction just above cloud base and a shallower cumulus layer, likely due to the inability to resolve finer turbulent eddies.
  • Resolution Threshold: The study suggests that a grid resolution of approximately Δx=250m and Δz=100m is likely necessary to realistically simulate subtropical boundary layer cloud feedbacks.

Limitations and Future Work

  • Computational Cost: Running CRMs at very fine resolutions (LES scales) is computationally expensive, limiting the ability to perform long climate simulations or to apply these resolutions globally in GCMs.
  • Stratocumulus Regimes: The study primarily focused on trade cumulus regimes (LTS70-80 and LTS80-90). The behavior of stratocumulus clouds (found in higher LTS bins, e.g., LTS90-100) in a warming climate, which are even more sensitive to resolution, was deferred to future work due to the complexity of simulating them accurately.
  • Diurnal Cycle: While the study explored the impact of a diurnal cycle of insolation, further investigation into its nuanced effects on cloud feedback is warranted.
  • Future Work: The authors plan to extend this column modeling approach to stratocumulus regimes and deep convection. They also aim to explore the implications of these findings for the design of future superparameterized GCMs, potentially requiring higher resolution embedded CRMs.

Summary updated

Abstract

A cloud resolving model (CRM) is used to investigate the low‐cloud increase due to a 2 K SST increase in the SP‐CAM superparameterized climate model. Of particular interest is the sensitivity of cloud changes to the CRM resolution (4 km horizontal, 30 vertical levels) in SP‐CAM. The CRM is run in column‐modeling framework using SP‐CAM composite cloud regimes. The experiments are run to steady state using composite advective tendencies, winds, and sea‐surface temperature from the control and +2 K climates of SP‐CAM. A new weak temperature gradient algorithm based on an idealized form of gravity wave adjustment is used to adjust vertical motion in the column to keep the simulated virtual temperature profile consistent with the corresponding SP‐CAM composite profile. Humidity is also slowly relaxed toward the SP‐CAM composite above the boundary layer. With SP‐CAM grid resolution, the CRM shows +2 K low cloud increases similar to SP‐CAM. With fine grid resolution, the CRM‐simulated low cloud fraction and its increase in a warmer climate are much smaller. Hence, the negative low cloud feedbacks in SP‐CAM may be exaggerated by under‐resolution of trade cumulus boundary layers.
BibTeX
@article{blossey2009subtropical,
  author = {Peter N. Blossey and Christopher S. Bretherton and M. C. Wyant},
  title = {{Subtropical Low Cloud Response to a Warmer Climate in a Superparameterized Climate Model. Part II: Column Modeling with a Cloud Resolving Model}},
  journal = {Journal of Advances in Modeling Earth Systems},
  year = {2009},
  doi = {10.3894/james.2009.1.8},
}
Figure