XII
III Simulating and Predicting Climate
119
7 The Simulation of Weather Types in GCMs: A Regional
Approach to Control-Run Validation
121
by KEITH R. BRIFFA
7.1 Introduction................
121
7.2 The Lamb Catalogue . . . . . . . . . . .
122
7.3 An "Objective" Lamb Classification ..
123
7.4 Details of the Seleded GCM Experiments
126
7.5 Comparing Observed and GCM Climates
128
7.5.1 Lamb Types . . . . . . . . . . . .
128
7.5.2 Temperature and Precipitation . .
131
7.5.3 Relationships Between Circulation Frequencies and
Temperature and Precipitation
133
7.5.4 Weather-Type Spell Lengths
and Storm Frequencies .
133
7.6 Conclusions.........
136
7.6.1 Specific Conclusions . .
136
7.6.2 General Conclusions . .
138
8 Statistical Analysis of GCM Output
139
by CLAUDE FRANKIGNOUL
8.1 Introduction..........................
139
8.2 Univariate Analysis. . . . . . . . . . . . . . . . . . . . . .
140
8.2.1 The t-Test on the Mean of a Normal Variable. . .
140
8.2.2 Tests for Autocorrelated Variables
141
8.2.3 Field Significance . . . . . . . . . . . . .
143
8.2.4 Example: GCM Response
to a Sea Surface Temperature Anomaly
143
8.3 Multivariate Analysis. . . . . . . . . . . . . .
145
8.3.1 Test on Means
of Multidimensional Normal Variables
145
8.3.2 Application to Response Studies
146
8.3.3 Application to Model Testing
and Intercomparison . . . . . . .
152
9 Field Intercomparison
159
by ROBERT E. LIVEZEY
9.1 Introduction................
159
9.2 Motivation for
Permutation and Monte Carlo Testing
160
9.2.1 Local vs. Field Significance
161
9.2.2 Test Example . . .
164
9.3 Permutation Procedures .
166
9.3.1 Test Environment
166