Interannual Variability of North American Summer Precipitation in NASA/NSIPP and NCAR/CAM2.0 AMIP Simulations

1
Interannual Variability of North American Summer
Precipitation in NASA/NSIPP and NCAR/CAM2.0 AMIP
Simulations

• By
• Alfredo Ruiz-Barradas and Sumant Nigam
• Department of Meteorology
• University of Maryland
• 
  September 3, 2003

2
Goal

• To assess interannual variability of
  precipitation over North America in AMIP-like
  runs of CAM2.0 and NSIPP models during summer
  months (June, July, August).

3
Data

• Precipitation
• Retrospective US and Mexico analysis.
• Hulme (University of East Anglia) data set.
• Xie/Arkin precipitation data set.
• SST from Hadley Center.
• NCAR/NCEP Reanalysis.
• AMIP simulation (ensemble no. 5) from the NSIPP
  model.
• AMIP simulation (case newsstamip06) from the CAM
  model.

4
Method

• Reanalysis and simulations extrapolated to a
  5?2.5? grid on 17 pressure levels.
• Monthly climatology for the 1950-1998 period.
• Monthly anomalies wrt 1950-1998 climatology.
• JJA is the mean of June, July, August.
• Assessment through
• Standard Deviation
• Precipitation Index
• Multivariate analysis

5
(No Transcript)
6
12 years peak
5 years peak
5 years peak
7
(No Transcript)
8
(No Transcript)
9
(No Transcript)
10
(No Transcript)
11
(No Transcript)
12
(No Transcript)
13
(No Transcript)
14
(No Transcript)
15
(No Transcript)
16
Remarks

• Large precipitation variability in observations
  and simulations over central US. Although it is
  shifted in simulations.
• Great Plains precipitation indices from
  simulations do not correlate with the observed
  index.
• SST regressions on the Great Plains index suggest
  linkage with Pacific midlatitude variability.
  CAM however emphasizes the tropical influence.

17
Remarks

• Multivariate analysis indicates
• Great Plains precipitation variability is the
  main mode of summer variability in observations
• This is however not the case in both model
  simulations
• Wet/dry events are cold/warm events in both
  observed and simulated summers.

18
Remarks

• PC regressions on moisture fluxes and
  geopotential heights indicate
• Observed precipitation variability is linked to a
  coherent, barotropic circulation that
  enhances/diminishes southerly stationary moisture
  flux from the Gulf of Mexico
• Model simulated variability does not have such
  circulation linkages.