Downscaling Tools

1
Downscaling Tools

• Introduction to LARS-WG and SDSM

2
LARS-WG stochastic weather generator (
http\\www.iacr.bbsrc.ac.uk\mas-models\larswg.html
)

• Generation of long weather time-series suitable
  for risk assessment
• Ability to extend the simulation of weather to
  unobserved locations
• A computationally inexpensive tool to produce
  climate change scenarios incorporating changes in
  means and in variability

3
LARS-WG stochastic weather generator(
http\\www.lars.bbsrc.ac.uk\model\larswg.html )

• Generates precipitation, min and max temperature
  and solar radiation
• Modelling of precipitation events is based on
  wet/dry series
• Semi-empirical distributions are used for
  precipitation amounts, dry/wet series and solar
  radiation
• Temperature and solar radiation are conditioned
  on the wet/dry status of a day
• Temperature and solar radiation are
  cross-correlated

4
LARS-WG

• Model calibration - SITE ANALYSIS
• Model validation - QTEST
• Generation of synthetic weather data - GENERATOR

5
SITE ANALYSIS
6
QTEST
Compare observed and synthetic data to evaluate
LARS-WG performance
7
GENERATOR
Generate synthetic weather data to extend time
series, or for climate change studies
8
GENERATOR
9
Limitations of LARS-WG (and weather generators in
general) ...

• Temporal downscaling only
• Designed for use at individual sites only (no
  spatial correlation)
• Can only represent events in calibration data set
• Generally underestimate variability

10
SDSM

1. A decision support tool for assessing local
   climate change impacts
2. Facilitates the rapid development of multiple,
   low-cost, single-site scenarios of daily surface
   weather variables under current and future
   climate forcing
3. Based on a multiple regression-based method

11
SDSM Structure

• 7 steps
• Quality Control and Data Transformation
• Screening of Predictor Variables
• Model Calibration
• Weather Generation (using observed predictors)
• Statistical Analyses
• Graphing Model Output
• Scenario Generation (using climate model
  predictors)

12
(No Transcript)
13
(No Transcript)
14
(No Transcript)
15
Model Verification
16
(No Transcript)
17
(No Transcript)
18
(No Transcript)
19
(No Transcript)
20
Tmax gt 25C
21
(No Transcript)
22
Cautionary Remarks

• SDSM provides a parsimonious technique of
  scenario construction that complements other
  methods
• SDSM should not be used uncritically as a black
  box (evaluate all relationships using
  independent data)
• Local knowledge is an invaluable source of
  information when determining sensible
  combinations of predictors
• Daily precipitation amount at individual stations
  is the most problematic variable to downscale
• The plausibility of all SDSM scenarios depends on
  the realism of the climate model forcing
• Try to apply multiple forcing scenarios (via
  different GCMs, ensemble members, timeslices,
  emission pathways, etc.)

23
Projet FACC (en cours 2003-2004)Etude sur
force/faiblesse de SDSM et LARS-WGpour extrêmes
et variabilité climatique
Coordonnateur Philippe Gachon Collaborateurs
- Ouranos Alain Bourque, René Roy, Claude
Desjarlais, Georges Desrochers, Vicky Slonosky,
Diane Chaumont - EC-SMC (Qc) Jeanna Goldstein,
Jennifer Milton, Nicolas Major - McGill VTV
Nguyen, Charles Lin - INRS-ETE André St
Hilaire, Bernard Bobée, Taha Ouarda - UQAM
Peter Zwack - CCIS Elaine Barrow - Post-Doc et
étudiants Tan Nguyen (PostDoc) Massoud Hessami
(PostDoc) Mohamed Abul Kashem (PhD)
24
1st Objective intercompare SDSM LARS-WG for
downscaling extremes (regional case-studies)
5 Régions à étudier (Stat. Downscaling)
1961-1990 Tmin Tmax Tmoy Precipitation tot.
2
1
4
3
5
25
(No Transcript)
26
2nd Objective Develop observed climate indices
used for verification analysis (using STARDEX
software)
27

THANK YOU FOR YOUR ATTENTION !!