The analysis based on the CYGP has been applied on the coupled GCM simulations prepared for the Intergovernmental Panel of Climate Change (IPCC) Fourth Assessment (AR4), and made available by the Program for Climate Model Diagnosis and Intercomparison (PCMDI1). For the analysis of the future climate change we have chosen to use the simulation A2 of the Special Report on Emission Scenarios (SRES), which is the economic scenario giving one of the larger increases of greenhouse gas (GHG) forcing at the end of the 21st century. The models for which results for scenario A2 were available and used in the present analysis are summarized in Table 1.

We have also used as validation for the current climate the ECMWF ERA40 reanalysis dataset (Uppala et al. 2005). The ERA40 dataset has been interpolated to a T42 resolution, which corresponds to about an average of the model resolutions. As another measure of model resolution, in addition to the standard descriptions in term of spectral truncation (indicated by the standard notation of T for triangular truncation and the maximum wavenumber), or grid spacing (in degrees of longitude

Model |
Resolution |
Number of ocean points |
Convective precipitation (mm/day) |
Calibration coefficient b |

INMcm3_0 |
5° x 4° |
892 |
3.02 |
0.256 |

GISS_model_e_r |
5° x 4° |
909 |
2.71 |
0.514 |

CCMA_cgcm3_1 |
T47 |
1240 |
2.14 |
0.603 |

IPSL_cm4 |
3.75° x 2.5° |
1988 |
2.15 |
0.429 |

UKMO_hadcm3 |
3.75° x 2.5° |
1995 |
3.34 |
0.335 |

MRI_cgcm2_3_2_a |
T42 |
2466 |
2.59 |
0.384 |

MIROC3_2_medres |
T42 |
2466 |
1.59 |
0.848 |

BCCR_bcm2_0 |
T63/G42 |
2478 |
3.36 |
0.822 |

CNRM_cm3 |
T63/G42 |
2480 |
3.60 |
0.713 |

GFDL_cm2_0 |
2.5° x 2° |
3807 |
2.95 |
0.406 |

CSIRO_mk3_0 |
T63 |
5344 |
2.42 |
0.620 |

MPI_echam5 |
T63 |
5396 |
3.01 |
0.328 |

UKMO_hadgem1 |
1.875° x 1.25° |
7994 |
3.48 |
0.346 |

NCAR_ccsm3_0 |
T85 |
9443 |
2.79 |
0.434 |

INGV_echam4 |
T106 |
14672 |
1.99 |
0.558 |

ERA40 |
T42 |
2480 |
2.27 |
0.543 |

1 see information at: http://www-pcmdi.llnl.gov/ipcc/data_status_tables.htm

and of latitude), we have also indicated the total number of grid points over ocean in the latitude belt 35°S-35°N, over which the TC cyclogenesis index is computed.

The models have been ranked in the order of increasing number of ocean grid points. The small differences in the number of ocean grid points in models with similar resolutions is mainly due to the use of slightly different land masks. The average annual convective precipitation over these ocean points is given in the fourth column of the table since it is a fundamental component of the CYGP. The last column gives the calibration coefficient p of the CYGP that is applied to each model.

It can be seen that the model resolutions encompass a range of resolutions of over an order of magnitude in the number of ocean points (from less that 900 points to more than 14000 points). The convective precipitation varies also within large limits, from less than 2 mm/day to more than 3 mm/day, without any apparent link to model resolutions. This variation can be due to the fact that models use very different convective parameterization schemes and have different biases in SSTs, which can influence the surface heat and moisture fluxes. It has been noticed that many models produce too much convective precipitation and too little stratiform precipitation compared to observational estimates (Dai 2006). In view of this wide variation we have chosen to make the CYGP convective precipitation threshold (PT) proportional to the mean convective precipitation of each model. The coefficient of proportionality (1.32) was chosen after determining that a threshold of 3 mm/day seemed suitable for the ERA40 reanalysis that has a mean convective precipitation of 2.27 mm/day. We have checked that in the models the results are not sensitive to the choice of a threshold, but for consistency with ERA40 we have kept the use of the threshold in the models as well.

Since the large variation in convective precipitation in the different models is largely due to their use of different parameterization schemes for convection, it seems appropriate to choose the calibration coefficient p used in the computation of the CYGP as model dependent. This calibration coefficient was chosen so that the total cyclogenesis over all the ocean points computed over the interval 1960-1999 was equal to 84 TCs per year, which is about the average number of observed cyclones. It can be seen that this calibration coefficient varies also in a large interval (from less than 0.3 to over 0.8), but not always in the opposite direction as convective precipitation, since dynamical factors may also modulate the CYGP.

Was this article helpful?

This is a collection of 3 guides all about survival. Within this collection you find the following titles: Outdoor Survival Skills, Survival Basics and The Wilderness Survival Guide.

## Post a comment