Zj fc

where Af. represents change in probability of prevalence of a vegetation type j. The V index always lies between 0 and 1, with 1 being complete change in vegetation composition and 0 no change. Figure 6.2 shows distribution of the V index computed from the observed fractions of vegetation and those estimated with probability density function under current climate. Figure 6.2 illustrates that

V Indn. m • -loi V* otmrvmlion*

■ OJitaO-a nil VO.IStoCZI tzsi

□ o K C 17 (3134| Figure 6.2. Comparison of the predicted vegetation composition under current climate to the observed vegetation composition.

Table 6.2. Range of changes in the climatic variables under four different climate scenarios.

Variable GISS GFDL CCC MM4

Table 6.2. Range of changes in the climatic variables under four different climate scenarios.

TMIN, °C

2.5 to 7.2

2.3

to 7.1

2.1

to

7.8

0.5

to

7.9

TMAX, °C

2.2 to 7.0

1.9

to 7.9

2.4

to

4.9

0.5

to

5.1

TMEAN, °C

3.1 to 5.3

3.0

to 6.6

2.2

to

5.9

2.3

to

5.3

BIOT,°C

1.7 to 5.3

1.6

to 5.4

1.0

to

5.4

1.3

to

4.1

PMEAN, mm/mn

-7 to 49

-19

to 63

-17

to

19

-19

to

434

Pwinter, mm/mn

-21 to 77

-72

to 16

-36

to

21

-32

to

231

Pspring, mm/mn

-15 to 48

-41

to 83

-26

to

87

-52

to

176

Psummer, mm/mn

-5 to 48

-57

to 240

-21

to

52

-83

to

471

Pautumn, mm/mn

-35 to 73

-15

to 86

-23

to

41

-50 to

1424

under current climate conditions the composition of vegetation is estimated very accurately in the majority of grid cells.

The estimated density functions then were used to calculate the probability of occurrence for each of the 11 vegetation types for each 1 degree by 1 degree grid cell and each of the four scenarios of climate change. Inland water bodies and the extent of agricultural and urban areas were considered constant under all scenarios. The climate changes were modeled as changes in each of the average and seasonal climatic variables. The new sets of climates were derived from VEMAP Climate Change Scenario Databases.The new sets of climate variables for each scenario were recomputed from the monthly temperature and precipitation variables obtained from three global circulation models (GISS, GFDL, and CCC GCMS) and one mesoscale atmospheric model (MM4) under assumptions of 2 X CO2 atmospheric concentration. The ranges of changes in each of the climatic variables are shown in Table 6.2. Additionally, Figures 6.3A, B and 6.4A, B display spatial patterns of changes in average temperature and precipitation. The table and figures indicate that all four scenarios predict a similar magnitude of warming, but the spatial patterns of warming differ markedly. The models' predictions vary in both the magnitude and the spatial distribution of the expected changes in precipitation.

The distributions of the V index for the four different climate change scenarios are presented in Figures 6.5A and 6.5B. Light gray indicates little or no change, and dark gray indicates substantial changes in composition of vegetation. In all four cases the major changes in composition of vegetation occur along the West Coast, in the Rocky and Appalachian Mountains, and in the Northeast of the United States. The Midwest and parts of the Southwest do not exhibit strong changes in vegetation composition. This result could be attributed to the fact that these regions are in agricultural or urban areas and were assumed to be unchanged. Changes also appear to be small in the lower part of the Southwest (e.g., Texas). However, in this case the small V index can be attributed to the fact that all four climate change scenarios show relatively small warming there without drastic reductions in precipitation. The West Coast and Rocky Mountains exhibit significant changes in composition of vegetation under all four scenarios. In the Northeast, Southwest, and Midwest, the most pronounced changes in vegetation composition occur under the GFDL scenario and the least under the CCC scenario. Table 6.3 illustrates the magnitude of possible changes in the total areas dominated by different vegetation types. Although these estimates were computed at two different climate-vegetation equilibria and do not represent the actual transient responses of vegetation, they still show whether the areas occupied by a specific vegetation type will decrease or increase with changing climate. For

Figure 6.3a. Changes in average temperature.
Figure 6.3b. Changes in average temperature.

example, it is likely that under all four scenarios the areas currently occupied by such coniferous species as spruce, fir, larch, and hemlock will decrease. Figure 6.6 shows a possible direction of migrations for some vegetation types under GFDL and GISS climate change scenarios. Deciduous (e.g., oak) and coniferous (e.g., pine,

Figure 6.4a. Changes in average annual precipitation.

spruce) vegetation types show generally northward movement in their prevalence under the assumed changes in climate. However, the extent and spatial configuration vary depending on the magnitude of warming and precipitation regime.

Table 6.3. Changes in the Area of Different Vegetation Types (sq. km)

Dominant vegetation or land cover

GISS

GFDL

CCC

MM4

1. Crops

0

0

0

0

2. Water

0

0

0

0

3. Wetlands

5,598

37,272

9,244

8,371

4. Grasslands

11,095

155,650

108,262

-2,378

5. Shrubs

189,468

133,091

235,071

153,880

6. Oak, hickory, etc.

-114,757

527

-12,015

-39,096

7. Beech and maple

-114,380

-148,409

-85,055

-102,507

8. Birch, aspen, etc.

-169,568

-225,532

-138,341

-142,558

9. Spruce, fir, etc.

-37,458

-51,881

-54,281

-25,840

10. Pine

233,935

146,065

-28,997

199,386

11. Pinion juniper

48,079

-9,528

-159

-24,550

12. Tundra

-6,329

-4,008

-6,630

-5,279

13. Barren /Ice

-45,616

-32,099

-27,174

-19,014

JJfclJA PMÇAN. WC

0 0

Post a comment