Results for Scenario A2

To look more clearly at the trends of the different models that were apparent in Fig. 2, the change in the yearly number in TC genesis over the different basins between the last 3 decades of the 21st century with respect to those of the last 3 decades of the 20th century have been computed. The results, shown in Table 3, have been normalized by expressing the evolution as a percent change with respect with their numbers at the end of the 20th century (shown in Table 2), so as to show more clearly the behavior of the different basins independently of their large differences in total TC genesis.

There is a rather good consistency of the models for simulating an increase in TC genesis over the North Indian Ocean. Only one model (GFDL_cm2_0) simulates a small decrease, and another model (INGV_echam4) a negligible increase. Most models give increases above 10%, and three of them produce increases above 40% (INM_cm3, GISS_model_e_r, BCCR_bcm2_0). Over the South West Indian Ocean the consistency decreases as four models simulate a decrease in TC genesis (MRI_cgcm2_3_2_a, GFDL_cm2_0, UKMO_hadgem1, INGV_echam4), the decrease being rather strong (-22%) in the case of the INGV model. Two models still give increases above 40% in this region (MIROC, CSIRO).

The situation is similar over the North West Pacific where four models give a negative change (CCMA_cgcm3_1, MIROC3_2_medres, CNRM_cm3, INGV_ echam4). Over the South West Pacific and North East Pacific there is no consistency even in the sign of the response since respectively 6 and 7 of the models, that is nearly half of the models, give a decreasing response and the rest a positive response. Over the Atlantic ten of the models give a decreasing in TC genesis, but still five of them an increase (GISS_model_e_r,, UKMO_hadcm3, BCCR,

Table 3 Change in the cyclogenesis per year simulated by the models in scenario A2 for the end of the 21st century (2070-2099) with respect to the end of the 20th century (1970-1999) for the total numbers, and relative change for the different ocean basins (in percent)

Model

Total

NI

SWI

NWP

SWP

NEP

ATL

INMcm3_0

12.1

58%

26%

25%

10%

-5%

-21%

GISS_model_e_r

30.5

59%

11%

39%

38%

48%

67%

CCMA_cgcm3_1

-1.7

14%

23%

-12%

-10%

18%

-23%

IPSL_cm4

7.3

13%

18%

5%

19%

-4%

-11%

UKMO_hadcm3

10.8

13%

10%

15%

-10%

49%

21%

MRI_cgcm2_3_2_a

5.4

35%

-7%

10%

9%

-6%

-11%

MIROC3_2_medres

-1.6

23%

62%

-15%

-19%

0%

-4%

BCCR_bcm2_0

9.5

41%

28%

9%

1%

36%

2%

CNRM_cm3

9.0

12%

21%

-1%

10%

35%

-13%

GFDL_cm2_0

-3.3

-8%

-8%

8%

-16%

-6%

39%

CSIRO_mk3_0

15.5

33%

45%

23%

9%

-1%

32%

MPI_echam5

21.6

30%

16%

31%

20%

46%

-25%

UKMO_hadgem1

10.6

9%

-5%

49%

4%

-3%

-55%

NCAR_ccsm3_0

-1.6

31%

8%

0%

-6%

-22%

-11%

INGV_echam4

-11.7

2%

-22%

-8%

-20%

0%

-43%

Mean

7.5

24%

15%

12%

3%

12%

-4%

S. Dev.

10.6

19%

21%

19%

17%

24%

32%

GFDL, CSIRO). From the dispersion of the model results it seems rather hazardous to draw a conclusion on the changes in TC genesis in the different ocean basins.

We can try to investigate further the cause of this dispersion by looking at the spatial distribution of the CYGP response. The changes of cyclogenesis in the different basins have a rather complex spatial distribution. This can be illustrated by the map of differences of the CYCP at the end of the 21st century compared to the end of 20th century (Fig. 3). It can be seen that the patterns of the cyclogenesis changes differ from model to model and have a rather complex structure with both positive and negative regions, which are an indication of displacements of the cyclogenesis areas. Though at first sight it may seem rather difficult to distinguish some coherency in model responses, a more detailed analysis reveals a number of general features.

The distribution of the response in longitude varies markedly according to the model and ocean basin. Over the Atlantic the response is very small except for the GISS_model_e_r, which gives a zonally symmetric increase over all ocean basins. This weakness of the response should, nevertheless, be related to the underestimation of the Atlantic TC activity as it is diagnosed by the CYGP in the models. Over the North Indian Ocean nearly all models simulate an increase of TC genesis west of India over the Arabian Sea. However, east of India over the Gulf of Bengal some models simulate an increase of TC genesis (BCCR, CSIRO, GISS, INM, MRI), while several other simulate a decrease (CCMA, CNRM-cm3, INGV, NCAR). This

Fig. 3 Spatial distribution of the model CYGP change at the end of the 21st century (2070-2099) with respect to the end of the 20th century (1970-1999). The unit is the number of TC genesis per 20 years per 5° x 5°. Positive contours are shown by solid lines and negative contours by dashed lines. The model identifier is shown above each figure

Fig. 3 Spatial distribution of the model CYGP change at the end of the 21st century (2070-2099) with respect to the end of the 20th century (1970-1999). The unit is the number of TC genesis per 20 years per 5° x 5°. Positive contours are shown by solid lines and negative contours by dashed lines. The model identifier is shown above each figure

110 11 Fig. 3 (Continued)

contrasting response in the Bay of Bengal explains the difference in the magnitude of the change over the North Indian domain shown in Table 3. A similar contrast between west and east is found in the South Indian Ocean, where nearly all models give an increase of TC genesis in the western part, while in the eastern part (east of

90°E) most models give a decrease (BCCR, CNRM, CSIRO, MIROC, MRI, NCAR, UKMO_Hadcm3), with only a few giving an increase (CCMA, GISS, IPSL, MPI). Two models have a different pattern of response in this area: the INGV model that gives a general decrease over this area, and the UKMO_hadgem1 that gives a zonal response with an increase equatorward and a decrease poleward, corresponding to a general equatorward migration of the TC genesis area over the South Indian Ocean.

It can be noticed that the same pattern of response extends over the South West Pacific in these two models. For several of the other models the response over the SWP basin shows a contrast in longitude with a positive response westward of about 180-150°W, and a negative response eastward (BCCR, CNRM-cm3, INMcm3, MPI_echam5, NCAR_cssm3, UKMO_hadcm3). This pattern of response corresponds to a reinforcement of TC genesis over the Coral Sea, and a reduction of the eastward extension of the cyclogenesis zone associated with the SPCZ.

A similar east-west contrast over the North Pacific is found only in a few models (CSIRO_mk3, INMcm3, IPSL-cm4, MRI_cgcm2_3_2a). The majority of models give rather a more zonally uniform response extending across the whole North Pacific (BCCR, CNRM-cm3, GFDL_cm2_0, GISS_model_e_r, MPI_echam5, NCAR_ccsm3_0, UKMO_hadcm3, UKMO_hadgem1) with most of the time a north-south contrast. In most of the models there is an increase of TC genesis between 10-20°N latitude situated on the poleward side of the main cyclogenesis area. Depending on the model this zone of increase can extend over a whole ocean basin or be restricted to a part of it. This pattern corresponds to a poleward extension of the TC genesis, which is consistent with the poleward expansion of the Hadley circulation that has been diagnosed in the IPCC AR4 climate simulations (Lu et al. 2007).

There are however some exceptions to this general pattern, for example MIR-OC3_2_medres that simulates an increase equatorward and a decrease poleward of the main cyclogenesis zone in the North Pacific, both in the west and in the east, as well as in the South West Pacific. A similar opposite pattern is also found in the NCAR_ccsm3_0 simulation. For CCMA this opposite pattern is found only over the West Pacific in both hemispheres, and for INGV only in the South West Pacific and South Indian Ocean. For these models this type of response explains the strong reduction in TC genesis in these basins noticed in Table 3.

In conclusion there appears a diversity of patterns of TC genesis changes in the different models, but for several models there is a similar response pattern in longitude leading to a displacement of TC genesis westward in the North Indian Ocean, South Indian Ocean and South West Pacific. This pattern could be associated to a response of the Walker circulation to global warming, as described by Vecchi and Soden (2007). As the response of the tropical circulation is very sensitive to the SST gradients, a possible explanation of this diversity of response patterns could be linked to the model SST response to GHG increase.

Indeed it can be seen that the SST changes differ considerably from model to model (Fig. 4) not only in their magnitude, but also in their geographical pattern. Several models have a pattern of warming showing a marked resemblance to an El

Fig. 4. Change in SST simulated by each model for scenario A2 at the end of the 21st century (2070-2099) with respect to the end of the 20th century (1970-1999). The contour interval is 0.5 K above 2K. The model identifier is shown above each figure
Fig. 4 (Continued)

Nino type anomaly with a maximum warming in the equatorial East Pacific (CCMA, CNRM_cm3, CSIRO, INGV, IPSL, MPI_echam5, UKMO_hadcm3) while others have a maximum warming in the central or West equatorial Pacific (INMcm3, MIROC, NCAR_ccsm3, UKMO_hadgem1). Such patterns of El Ninolike or La Nina-like responses to global warming have been previously pointed out (Collins et al. 2005, Yamaguchi and Noda 2006). It can be expected that the different patterns of SST changes in the model can modify the location of convec-tive rainfall and also the other circulation features, such as wind shear, that can have a controlling effect on TC genesis (Vecchi and Soden 2007, Latif et al. 2007).

The lack of agreement in the pattern of SST changes can be considered to be one of the principal factors that is responsible for the diversity of the patterns of change in TC genesis.

Was this article helpful?

0 0
Survival Treasure

Survival Treasure

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.

Get My Free Ebook


Post a comment