This research team will formulate and apply a simplified numerical model to explore conditions favoring the self-aggregation of moist atmospheric convection. Aggregation of cumulus convection into clumps containing many individual convective cells is prominent in nature and in full-physics numerical atmospheric simulations alike. The researchers have performed preliminary analyses using the System for Atmospheric Modeling (SAM) numerical model, and will expand upon these simulations in order to address the hypothesis that self-aggregation of convection will lead to a systematic decline in the surface temperature, and that such decline may in-turn lead to disaggregation of convection and restorative re-warming of the surface. As such, the system would naturally be attracted to the phase transition between aggregated and disaggregated states and would constitute an example of a self-organized critical system. The objectives of the proposed work are to: (1) understand the physics of self-aggregation, (2) determine how much hysteresis there is in the system, (3) determine whether radiative-convective equilibrium systems with surface energy balance are subject to self-organized criticality, and (4) understand the implications of these aforementioned factors upon climates predicated on idealized radiative-convective states as well as more real-world conditions.
The intellectual merit of the work derives from improved basic understanding of mechanisms influencing moist atmospheric convection, which is a critical but nonetheless not fully understood component of the climate system. A confirmation of the hypothesis that tropical climate tends toward a state of self-organized criticality could substantially advance the understanding of the interrelationship between convection and climate.
The broader impacts of the study include the education and training of a graduate student at each of two collaborative institutions. The work also has potential to lead to improved representations of convection via improved parameterizations suitable for incorporation into coupled climate models, which could in turn contribute to improved climate forecasts.
Hurricanes are among the most destructive natural phenomena on this planet. Powerful tropical cyclones such as hurricanes Katrina and Candy can cause tragic loss of lives and exert tremendous damage to the infrastructure. One of the important problems in climate research is whether the global warming due to increase of greenhouse gases in the atmosphere can affect the intensity and frequency of hurricanes. The research conducted in this project is a step in the direction of better understanding the link between the hurricanes and the warming Tropics. We use a cloud-resolving model, which is the tool to simulate tropical cyclones in great details. A cloud resolving model is a computer model that solves numerically the equations describing the three-dimensional evolution of the cloudy atmosphere over a domain big enough to contain many clouds or several tropical cyclones. To avoid including myriads of details that may not be of the first-order importance to our problem, we use an idealization for the tropical atmosphere called the radiative-convective equilibrium (RCE). In RCE framework, the clouds and hurricanes evolve in response to computed surface fluxes and radiative fluxes in the atmosphere. The surface fluxes depend locally on the sea-surface temperature (SST) and wind. The solar and infrared thermal radiative fluxes are also computed by the model using sophisticated radiative transfer scheme. The RCE simulations are expensive as they need to be run for at least 100 days to collect sufficient statistics about the cloud fields and other mean properties of the simulated atmosphere. There is a general consensus among scientists that the RCE framework is a great proxy for the tropical atmosphere. The novel aspect of this project is that the RCE has not previously been used to study the statistics of multiple tropical cyclones using a cloud-resolving model. One of the main reasons is very large computation expense in terms of computing resources of running the CRM over the domain big enough to contain several tropical cyclones evolving simultaneously over 100-day period. We need many cyclones at once because one of the main objectives of this study is to see how the number of cyclones in RCE can change in response to the change of the SST. Although we have used a powerful supercomputer for this project, to make the problem computationally feasible, we needed to increase the number of hurricanes per domain making them smaller by increasing planetary rotation and associated Coriolis force. Making hurricanes smaller does not change the fundamental physics of simulated hurricanes or their interaction with the surface and the surrounding atmosphere. To see the response of hurricane statistics such as intensity and frequency of occurrence to change in SST, we ran several 100-day simulations systematically varying temperature from 21 to 36 degrees Celsius. Figure 1 shows the snapshots of hurricanes in the numerical domain visualized by the surface pressure for different surface temperatures. One can see that the number of hurricanes tends to decrease with increasing SST. Despite the decrease of the number, the size and intensity of the hurricanes tend to increase with increasing SST. The per-hurricane kinetic energy of the wind and total precipitation rate tends to approximately double for a 6 degrees Celsius rise of the SST. In addition to simulations, we developed a quantitative theory that explains quite well the simulated change of number and intensity of hurricanes in response to the SST change. The results of this research have been published in one of the AGU's journals (see below). By the time of writing this report, the article has been one of the most accessed articles from the journal's website. The article has also been highlighted in AGU's EOS publication (March 25th 2014 issue). The results have also been reported at major conferences such as the AMS Conference on Hurricanes and Tropical Meteorology. Publication Khairoutdinov. M.K., and K. A. Emanuel (2013). Rotating radiative-convective equilibrium simulated by a cloud-resolving model. Journal of Advances in Modeling Earth Systems. 5 (4), 816. DOI: 10.1002/2013MS000253 Conference Papers and Presentations Khairoutdinov (2011). Climate feedbacks in idealized radiative-convective equilibrium simulations with tropical cyclones.. CFMIP/GCSS/EUCLIPSE Meeting on Cloud Processes and Climate Feedbacks. The Met Office, Exeter, United Kingdom. Khairoutdinov (2013). Organization of tropical convection and equilibrium climate sensitivity. GASS/MJO-TF meeting on diabatic processes of the MJO. Singapore. Khairoutdinov and Emanuel (2013). Response of tropical cyclones to warming SST in radiative-convective equilibrium with rotation. EUCLIPSE - CFMIP meeting. Paris. Khairoutdinov and Emanuel (2012). The effects of aggregated convection in cloud-resolved radiative-convective equilibrium.. 30th AMS Conference on Hurricanes and Tropical Meteorology. Jacksonville, FL. Khairoutdinov (2011). Updates, Tropical Cyclones.. NSF Center for Multiscale Modeling of Atmospheric Processes 11th Team Meeting,. Fort Collins, CO.
