On the Nature of the Recent Southwestern Drought
































- Slides: 32
On the Nature of the Recent Southwestern Drought Philip Pegion 1, Siegfried Schubert, Randy Koster, Max Suarez. NASA-GSFC Global Modeling and Assimilation Office 1. SAIC
Motivation We want to understand the cause of the recent drought, and can we determine if it has ended? Characteristics of recent drought • Started with La Nina in 1999 • 5 years in a row of below normal precipitation • 2002 was 4 th driest year on record, the region received only 70% of normal precipitation, and in 2003, the region only received 83%. • Driest period since the 1950 s. • Persisted past weak El-Nino in 2002
Desert SW. region YEAR
Observed Precipitation Desert SW region. Anomaly mm/month 25 -month running mean anomaly (mm/day) YEAR
Observed Precipitation Desert SW region. Anomaly mm/month 25 -month running mean anomaly (mm/day) YEAR
Observed Precipitation: % of normal
HADISST Sea Surface Temperature anomaly (K)
Observed Precipitation Desert SW region. ENSO Anomaly mm/month 25 -month running mean anomaly (mm/day) 2 nd EOF: HADISST Correlation: 0. 51 YEAR
Model Description • NASA Seasonal-to-Interannual Prediction Project (NSIPP-1) AGCM – – – • AMIP – – – • 4 th Order finite difference dynamical core (Suarez and Takacs, 1995). Simple K-scheme boundary layer (Louis et al. , 1982). Relaxed Arakawa-Schubert (RAS) convection (Moorthi and Suarez, 1992). Chou and Suarez (1994, 2000) radiation. Mosaic land model (Koster and Suarez, 1996). 3. 75 o longitude, 3 o latitude, 34 levels 14 member ensemble (1902 -present) Additional AMIP runs at 2 o and 1 o degree. DSP – – 0. 625 o longitude, 0. 5 o latitude, 34 levels Initialized December 1, 2001 with the NCEP Reanalysis.
Precipitation Desert SW region. 25 -month running mean anomaly (mm/day)
Precipitation Desert SW region (1998 -2004) 7 -month running mean anomaly (% of normal) CMAP Model ensemble mean
October 2001 -March 2002 Precipitation (% from normal)
October 2001 -March 2002 Surface Air Temperature Anomaly (K)
October 2001 -March 2002 200 mb Height Anomaly (m)
October 2001 -March 2002 200 mb Height Anomaly (m) Model La NINA Composite NCEP La NINA Composite
October 2001 -March 2002 850 mb v’q’ Anomaly (m s-1 g Kg-1)
High Resolution GCM runs • Initialized 00 z 1 DEC 2001 with the NCEP/NCAR reanalysis. • Run through end of March. • 10 ensemble member run over Reynolds SST • 10 additional ensemble members with MODIS SST.
DJFM 2002: SST anomaly (K) MODIS Reynolds MODIS-Reynolds
DJFM 2002: Precipitation Anomaly (mm/day) 0. 5 deg DSP w/ MODIS SST CMAP 0. 5 deg DSP w/ Reynolds SST MODIS - Reynolds
DJFM 2002: Precipitation (% of normal) 0. 5 deg DSP w/ MODIS SST CMAP 0. 5 deg DSP w/ Reynolds SST MODIS - Reynolds
DJFM 2002: 200 mb Height Anomaly (m) 0. 5 deg DSP w/ MODIS SST 0. 5 deg DSP w/ Reynolds SST NCEP/NCAR Reanalysis MODIS - Reynolds
DJFM 2002: Precipitation (% of normal) Ensemble 1 MODIS SST Ensemble 2 MODIS SST 0. 5 deg DSP w/ Reynolds SST Ensemble 2 Reynolds SST NCEP/NCAR Reanalysis
Has the long term drought ended, or is last winter just a bump in the time series?
Annual Mean SST Anomaly (K)
Annual Mean SST Anomaly (K)
12 month coupled model forecast
Conclusions • Start of drought is linked with changes in Pacific Ocean SSTs • Most severe part of drought, 2002, lies outside of the envelope of response in the model. • Some hi-resolution runs do get the correct magnitude of the winter drought. • Last winter rains provided big relief to the drought, but the southwest hasn’t recovered yet. Surface Elevation • Recent drought was of the most severe and longest lasting in recent times. Lake Powell Year
JFM 2002: 200 mb Height Anomaly (m)
DJFM 2002: 200 mb Height Anomaly (m)
DJFM 2002: 200 mb Height Anomaly (m)
Precipitation Desert SW region (1998 -2004) 7 -month running mean anomaly (mm/day) CMAP Model ensemble mean
Precipitation Desert SW region (1998 -2004) 7 -month running mean anomaly (normalized) CMAP Model ensemble mean