¥Minimum 4 main IPCC scenarios (A1, A2,
B1, B2), about 20 global models, 6-member GCM ensemble, 10 regional models, 8 ensemble members per regional model.
¥4x20x6x10x8 = 38,400 regional climate
model runs (or 3,840 runs per regional model). Not practical!
¥What are the greatest sensitivities in
nested global-regional climate models?
¥How can we most efficiently employ
our computer time and (most important) people?
¥Results from ensemble studies using GCMs
and short-range forecast models may not apply:
¥Regional climate models are (should be)
constrained by lateral boundary conditions.