PIs: Thomas R. Knutson and Robert E. Tuleya
NOAA/GFDL
P.O. Box 308
Princeton, NJ 08542
Phone:
Fax:
Email: tk@gfdl.noaa.gov, rt@gfdl.noaa.gov
One concern about future climate change is that the intensity of tropical cyclones could increase due to the warmer SSTs predicted by coupled climate models forced by increasing greenhouse gases. To explore this problem, the GFDL hurricane prediction system, a regional triply nested moveable mesh hurricane model with initialization system, has been utilized in combination with GFDL global climate model data (Knutson and Tuleya 1999). A set of idealized hurricane experiments have been performed where a weak tropical storm is initialized in a favorable tropical environment with uniform easterly flow (no wind shear or land) and allowed to intensity to a quasi-steady state intensity. The SSTs, atmospheric temperature, and atmospheric moisture environmental states for these high resolution experiments have been derived from climate experiments (control and high CO2) from a GFDL coupled model, using time and area means for various tropical storm basins. The storms simulated for the high CO2 environments typically have been slightly stronger (~5-10 percent higher maximum surface wind speeds) than for the control (present day) environments.
An important factor controlling hurricane intensity in these experiments (besides SST) is the vertical profile of the temperature changes throughout the troposphere. The enhanced warming in the upper troposphere, relative to the surface warming, acts to limit the amount of storm intensification simulated (Shen et al. 1999). Since only the GFDL coupled model environmental states have been used to date, the question arises as to what type of changes would be simulated using simulated future atmospheric/SST conditions from other climate models.
The objective of this subproject is to explore how sensitive the hurricane intensity simulation results of Knutson and Tuleya (1999) are to the choice of GCM used for the environmental state. Do the thermodynamic states simulated by various CMIP2+ models imply more intensification, less intensification, or even weakening of future hurricanes according to this methodology?
We propose to derive time- and area-averaged SST, air temperature, and moisture profiles from the participating CMIP2+ models over the various tropical storm basins and perform idealized hurricane model experiments with those environmental states.
Using the monthly climatologies of SST, temperature, and moisture for control and high CO2 conditions from the various models, the tropical storm basin environmental states and their CO2-induced changes will be obtained and compared graphically. Some CAPE or Maximum Potential Intensity (MPI) calculations may be done on these profiles. The main activity of the subproject will be to use these profiles as input for idealized experiments using the GFDL hurricane model following the methodology of Knutson and Tuleya (1999) although without ocean coupling (Knutson et al. 2001). The climate model-derived fields are used as reference boundary conditions at the central latitude of the east-west channel in which an idealized initial disturbance is placed in the hurricane model. The surface pressure and temperature fields for the hurricane model environment are computed at the end of the initial vortex insertion procedure by solving a form of the reverse balance equation. Aside from the disturbance vortex, the environmental flow will be specified as an easterly flow (~5m/sec) that is uniform with height. An ensemble (~10) of experiments will be run for each model/basin/CO2 condition in order to separate the actual simulated climate "signal" from internal variability "noise" in the model responses.
The hurricane model will likely be run using its current operational configuration in terms of resolution (maximum of about 18 km) and physics, although experiments using enhanced resolution or physics may be attempted.
The hurricane model simulations will be analyzed in particular with regard to maximum storm intensity (minimum central surface pressure or maximum surface wind speed) and precipitation (storm maximum or near- storm area average). Some evaluation of the hurricane model's simulated intensities has been done for "present climate" conditions (Knutson et al. 1998; Knutson and Tuleya 1999), although this remains an important research issue to address.
Knutson, T. R., and R. E. Tuleya, 1999: Increased hurricane intensities with CO2-induced global warming as simulated using the GFDL hurricane prediction system. Climate Dynamics, 15, 503-519.
Knutson, T. R., R. E. Tuleya, W. Shen, and I. Ginis: 2001: Impact of CO2-induced warming on hurricane intensities as simulated in a hurricane model with ocean coupling. Journal of Climate, 14 (11), 2458- 2468.
Shen, W., R.E. Tuleya, and I. Ginis, 2000: A sensitivity study of the thermodynamic environment on GFDL model hurricane intensity: implications for global warming. Journal of Climate, 13, 109-121.