New NWS Western Region New NWS Western Region Local Climate Products Local Climate Products 1 Marina Timofeyeva, 2 Andrea Bair and 3 David Unger 1 UCAR/NWS/NOAA 2 WR HQ/NWS/NOAA 3 CPC/NCEP/NWS/NOAA Contributors: Bob Livezey, Shripad Deo, Heather Hauser, Holly Hartmann, Eugene Petrescu, Michael Staudenmaier
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
New NWS Western Region New NWS Western Region Local Climate ProductsLocal Climate Products
1Marina Timofeyeva, 2Andrea Bair and 3David Unger
1 UCAR/NWS/NOAA
2 WR HQ/NWS/NOAA
3 CPC/NCEP/NWS/NOAA
Contributors: Bob Livezey, Shripad Deo, Heather Hauser, Holly Hartmann, Eugene Petrescu, Michael Staudenmaier
OUTLINEOUTLINE
• Need for Local Climate Products
• Challenges in Local Climate Product Development
• Methods and Data
• Product Design
• Operational Organization
• Next Steps
Need For Local Climate ProductsNeed For Local Climate Products
• CPC products and Local Climate
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0.16
0.18
24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54
CD83 SLC Ogden Heber Pl. Grove Logan
Need For Local Climate ProductsNeed For Local Climate Products
• Localized Climate Impacts are of public interest
Figures courtesy of Klaus Wolter, CDC
Challenges in Local Climate Challenges in Local Climate Product DevelopmentProduct Development
ScientificallyScientificallySoundSound
OperationallyOperationallyOrganizedOrganized
CustomerCustomerFriendlyFriendly
Local Local ClimateClimate
ProductsProducts
Methods and DataMethods and Data
• Modified CPC Translation of CD Seasonal Temperature POE
Forecasted Temperature (Forecasted Temperature (°F)°F)
PO
F (
%)
PO
F (
%) Observed T
Methods and DataMethods and Data
• Modification included:– Regression coefficients estimate: use of straight
regression coefficients versusversus ones inflated by correlation;
– Forecasting methodology: station mean and variance are estimated from CD forecasted mean and variance and use of normal distribution for POE ordinates versusversus use of inflated correlation coefficients and CD POE temperature ordinates;
– Local Product design is customer friendlier
Forecast issued in 09/2004 for JFM 2004 (3.5 lead)
0
20
40
60
80
100
37 39 41 43 45 47 49
Temperature (F)
PO
E (
%)
Forecast Climatology
Methods and DataMethods and Data
• Data: NCDC provided an experimental “homogenized and serially complete data” set with:– Monthly/daily value internal consistency
check– Bias adjusted to a midnight to midnight
observation schedule– Spatial QC– Artificial change point detected and adjusted– Estimated missing or discarded data
Methods and DataMethods and Data
0.5
0.6
0.7
0.8
0.9
1.0
0.5 0.6 0.7 0.8 0.9 1
0.5
0.7
0.8
0.9
1
ri – Station/CD Correlation
ρ (CD fcst/obs corr)
Sp
read
of
Sta
tion
Fore
cast Climatological Spread
Confident Prediction
Methods and DataMethods and Data
0
5
10
15
20
25
30
35
FMA MAM AMJ MJJ JJA JAS ASO SON OND NDJ DJF JFM
#of s
ites
with
cor
r>=
0.8
Regular NCDC data Special NCDC data
Methods and DataMethods and Data
• CPC Composite Analysis extended by Risk Analysis and CPC forecasting method
-5 0 5 10 15 20 25
0.0
0.0
50
.10
0.1
50
.20
0.2
50
.30
-5 0 5 10 15 20 25
0.0
00
.05
0.1
00
.15
0.2
00
.25
0.3
0
1941-2000
1941,1958,1966,1973,1983,1987,1988,1992,1995,1998
Eastern North Dakota Temperature (°F) Eastern North Dakota Temperature (°F)
Methods and DataMethods and Data
• Extension includes Risk Analysis identifying statistically significant signal
El Nino Above
0.0
0.1
0.2
0.3
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
# of events
Pro
bab
ilit
y
Methods and DataMethods and Data
• Making forecast using Composite Analysis
FORECAST USING CURRENT CPC Nino 3.4:FORECAST USING CURRENT CPC Nino 3.4:
Nino3.4
TermWarm Neutral Cold
Above 67% 33% 11%
Near 13% 53% 28%
Below 20% 14% 61%
P P P P P P Pca tegorysta tion
above even tsta tion
aboveN ino
near even tsta tion
nearN ino
below even tsta tion
belowN ino /
./
./
.* * *3 4 3 4 3 4
Example – ElNino with 7.5 month lead (forecast for JFM 2005):Example – ElNino with 7.5 month lead (forecast for JFM 2005):
NINO 3.4 INITIAL TIME 5 2004 PROJECTION FRACTION Lead Mo BELOW NORMAL ABOVEJJA 0.5 0.076 0.371 0.552…………………………………………DJF 6.5 0.053 0.388 0.559JFM 7.5JFM 7.5 0.0800.080 0.3930.393 0.5270.527