Diapositiva 1

1
A reconciling blocking index applied to
observations and GCMs
2
Perspectives
Perspectives
Concept Diagnosis
GCMs Conclusions
Objective design Homogeneization (same period
reanalysis) Update Aplication to SH GCMs
Blocking (Barriopedro et al. 2006)
Cyclones (Trigo et al. 1999)
WTs (Trigo DaCamara 2000)
Cut-off lows (Nieto et al. 2005)
Jet stream (Gallego et al. 2005)
Barriopedro et al. (2006). J. Clim. Trigo et al.
(1999). J. Clim. Trigo DaCamara (2000). Int. J.
Clim. Nieto et al. (2005). J. Clim Gallego et al.
(2005). Clim. Dyn.
3
Concept
Perspectives
Concept Diagnosis
GCMs Conclusions

• Large-scale, quasi-stationary and persistent
  anticyclonic systems that interrupt the regular
  eastward progression of the synoptic
  perturbations
• Blocks occurrence implies a strong modification
  of the westerlies ? significant impact in the
  regional climates of the mid and high latitudes
• Almost no studies addressing blocking in GCMs
  (IPCC, 2007), even when they were first mentioned
  at the beginning of the 20th century!!!

Trigo et al., (2004). Clim. Dyn. García-Herrera
et al. (2007). J. Hydromet.
4
Blocking indices
Perspectives
Concept Diagnosis
GCMs Conclusions

• Two main traditions based on absolute (Tibaldi
  and Molteni 1990, TM) and anomaly fields (Dole
  and Gordon 1983, DG)
• Each definition has their own assets and
  caveats, emphasizing different and partial
  blocking features ? a combined approach is
  expected to bring a more efficient diagnosis,
  avoiding shortcomings that arise from single
  approaches

Classification and evaluation of previous
reported blocking methods.
5
Blocking definition
Perspectives
Concept Diagnosis
GCMs Conclusions

• Blocking index a reconcinling methodology is
  designed by detecting 2-D Z500 positive anomalies
  above a given threshold (za) associated with
  easterly winds around a latitude (?c)
  representative of the jet stream

Scheme of daily blocking detection
6
Critical parameters
Perspectives
Concept Diagnosis
GCMs Conclusions

• Reference latitude the latitude of max.
  high-frequency (lt5 days) Z500 variance for
  running windows of 90-dy and 25-yr
• Anomaly threshold s (115 gpm).

Left Mean annual (solid) and seasonal (dashed,
dotted) reference latitude. Light (dark) shaded
indicates 1s of the monthly (annual) means.
Right High-pass Z500 variance (shaded), U500
(contours) and the reference latiude.
Annual distribution of daily Z500 anomalies and
the 1s values.
Series of daily Z500 (up) and Z500 anomalies
(bottom). 5-day persistent anomalies higher than
115 gpm are highlighted
7
Secondary criteria
Perspectives
Concept Diagnosis
GCMs Conclusions

• Extension criterion Am LR ?F 7.5º 15º
  106 km2
• Tracking A blocking anomaly detected in di is
  said to persist the next day di1 if it shows a
  minimum overlapping of 30. (equivalent to a
  maximum speed of 10º/day)
• Duration criterion t 5 days

Histogram of blocking event durations (right) and
exponential fit in a log-plot (left)
8
Comparison with previous indices
Perspectives
Concept Diagnosis
GCMs Conclusions
9
Application to GCMs
Perspectives
Concept Diagnosis
GCMs Conclusions

• Thresholds are objective and specific of the
  data set employed, being indicators of the model
  performance.
• Model experiments a) 25 time-slice members of
  40-yr from a 1000-year control simulation (CTL)
  and b) the ensemble (FOR) of two 1000-1989 forced
  simulations of the ECHO-G AOGCM (e.g.
  González-Rouco et al. 2008).

Critical parameters in OBS (dark) and ECHO-G
(light) reference latitude (left) and anomaly
threshold (right).
10
Model performance
Perspectives
Concept Diagnosis
GCMs Conclusions

• Preferred regions for blocking occurrence are
  well captured, but with a considerable
  underestimation over the Euro-Atlantic sector in
  cold seasons (NDJ) and Eurasia in warmer months
  (MJJ).
• Most of the frequency underestimation relates to
  the model inability to simulate long-lasting
  blocking events. However, the model provides a
  realistic simulation of the shape of the annual
  cycle.

2-D blocking frequency in OBS (lines) and FOR
(shaded).
Left Blocking duration distribution and
characteristic time scales in OBS (light) and FOR
(dark). Right annual cycle of the blocking
frequency in OBS (solid) and FOR (dotted), with
differences in the bottom.
11
Model errors
Perspectives
Concept Diagnosis
GCMs Conclusions

• The model failure arises from its inability to
  simulate height reversals over the Euro-Atlantic
  sector (i.e. biases in the absolute field),
  rather than anomalies of the right amplitude
  (i.e. biases in intra-seasonal variance).
• Two superimposed model errors 1) excessive
  zonal wind and 2) southward shift of the jet
  stream. These biases are observed trough the
  whole year, but they peak in NDJ over the
  Euro-Atlantic and in MJJ over Eurasia,
  respectively, being timely and coincidental with
  blocking underestimations.

Systematic model errors U500 (left) and location
and intensity of the jet stream (middle and right)
12
Conclusions
Perspectives
Concept Diagnosis
GCMs Conclusions

• An reconciling blocking index between
  traditional approaches is proposed, being
  applicable to observations and model simulations
  of different resolutions, temporal lengths and
  time variant basic states.
• The method succeeds in identifying persistent
  anomalies with blockage of the westerlies and
  avoids artificial assumptions and identification
  problems associated to other indices.
• The ECHO-G model captures with reasonable
  realism the main blocking features, namely
  preferred locations, annual cycle and exponential
  duration distributions, but reveals a
  considerable underestimation over the
  Euro-Atlantic sector.
• Two model biases are responsible for the
  blocking underestimation 1) its inability to
  generate amplified waves due to excessive zonal
  winds in cold months and 2) its tendency to place
  the jet stream further south of the observed one
  in warm months.
• As compared to the industrial period, Pacific
  blocking was more frequent in the past, whereas
  prolonged periods of high Atlantic blocking
  activity dominated the 17th and 18th centuries,
  supporting a blocking effect in the cold
  conditions that affected Europe during the
  Maunder Minimum.