Exotic Brome-Grasses in Arid and Semiarid Ecosystems of the Western US

392

and preserve potential seasonal and multiyear climate patterns in the data, the 1950
to 1999 time series was wrapped around as a time loop (i.e., year 1950 followed
year 1999) and was resampled by randomly selecting five start years (replicate
#1 = 1956; replicate #2 = 1954; replicate #3 = 1972; replicate #4 = 1960; replicate
#5 = 1990) until 75 years of data were accumulated. The next step was to use the
historic and projected CO 2 , temperature, and precipitation to create the final tempo-
ral multipliers.

01530456075

-6

-3

0

3

6

Replicate #1

PDSI

Climate Change A2

No Climate Change

01530456075

-6

-3

0

3

6

Replicate #2

0 15 30 45 60 75

-6

-3

0

3

6

Replicate #3

PDSI

01530456075

-6

-3

0

3

6

Replicate #4

Time Step (Year)

01530456075

-6

-3

0

3

6

Replicate #5

Time Step #5

PDSI

Fig. 13.6 Replicated and calculated PDSI time series using Eqs. (13.1) and (13.2) assuming no
climate change (resampled historic precipitation and temperature time series) and assuming the A2
scenario from five Global Circulation Models as five 75-year replicates. Climate change legend:
replicate #1 = mri_cgcm2_3_2a.2.sresa2; replicate #2 = ukmo_hadcm3.1.sresa2; replicate
#3 = ncar_pcm1.1.sresa2; replicate #4 = ncar_ccsm30.1.sresa2; and replicate #5 = ncar
ccsm3_0.2.sresa2

L. Provencher et al.