Seasonal Degree Day Outlooks

1
Seasonal Degree Day Outlooks

• David A. Unger
• Climate Prediction Center
• Camp Springs, Maryland

2
Definitions
_
_

• HDD G65 t t lt 65 F
• CDD G t 65 t gt 65 F
• HD HDD/N CD CDD/N
• T 65CD-HD
• CD T 65 HD
• t daily mean temperature, TMonthly or Seasonal
  Mean
• N Number of days in month or season

_
_
_
_
_
_
3
CPC Outlook
4
CPC POE Outlooks
5
(No Transcript)
6
Overview
Tools
Forecaster Input
Skill Heidke .10 RPS .02
Temperature Fcst Prob. Anom.For Tercile (Above,
Near, Below)
Model Skills, climatology
Temperature POE
Skill CRPS .03
Downscaling (Regression Relationships)
Temperature POE Downscaled
Skill CRPS .02
Temperature to Degree Day (Climatological
Relationships)
CRPS Skill CDD .05 HDD .02
Degree Days HDD CDD POE
Accumulation Algorithms
Degree Days Flexible Region, Seasons
CRPS Skill CDD .06 HDD .02
7
Temperature to Degree Days
8
Rescaling
Downscaling
FD Seasonal
CD Seasonal
Disaggregation
CD Monthly
FD Monthly
9
Downscaling

• Regression
• CD a FD b
• Equations coefficients are inflated
• (CD variance climatological variance)

10
Disaggregation - Seasonal to Monthly

• Tm a Ts b
• Regression, inflated coefficients
• Average 3 estimates
• M JFM M FMA M MAM
• 3

M
11
Verification note

• Continuous Ranked Probability Score
• - Mean Absolute Error with provisions
• for uncertainty
• Skill Score 1.
• - Percent Improvement over climatology

CRPS
CRPS
Climo
12
Continuous Ranked Probability Score
13
CRPS Skill Scores Temperature
.027 .029
.026 .023
.051 .045
.041 .034
.013 .016
.027 .026
.002 .001
.011 .004
-.009 .002
-.006 -.008
.040 .036
.026 .030
.055 .059
.055 .058
.044 .038
.050 .047
.020 .021
.024 .024
Skill
.035 .030
.012 .015
High
Moderate
Low
None
.094 .103
.074 .090
.10
.065 .055
.042 .035
.05
.01
FD
CD
.031 .023
.028 .019
3-Mo
1-Month Lead, All initial times
1-Mo
14
CRPS Skill Scores Heating and Cooling Degree Days
.114 .085
.019 .028
.040 .071
.036 .073
.058 .043
.021 -.011
.009 .022
.000 -0.16
-.004 .036
-.026 -.016
.101 .121
.014 .076
.090 .090
.029 .035
.035 .014
.045 -.003
.047 .102
.023 .048
Skill
.033 .051
.005 .003
High
Moderate
Low
None
.088 .115
.079 .111
.10
.044 .024
.046 .030
.05
.02
1-Mo
12-Mo
.049 .057
.018 .016
Cooling
Heating
15
Degree Day Forecast (Accumulations)
16
(No Transcript)
17
Reliability
18
Reliability
19
Conclusions

• Downscaled forecasts nearly as skillful as
  original temperature outlook
• Skill better in Summer than Winter
• Better understanding of season to season
  dependence will lead to improved forecasts for
  periods greater than 3-months.

20
Testing

• 50 years of perfect OCN
• Forecast decadal mean and standard
  deviation
• Target year is included to assure skill.
• Seasonal Forecasts on Forecast Divisions supplied
• How does the skill of the rescaled forecasts
  
• compare to the original

21
CRPS Skill Scores Downscaled and disaggregated
.098 .081
.061 .042
.088 .092
.063 .039
.086 .083
.061 .059
.088 .085
.061 .055
.108 .105
.061 .060
.106 .019
.067 .077
.110 .086
.066 .066
.074 .070
.052 .037
.109 .109
.058 .055
Skill
.138 .140
.086 .067
.198 .233
.106 .135
High
Moderate
Low
None
.10
.110 .087
.074 .044
.05
FD
CD
.01
.104 .109
.066 .057
Seasonal
Monthly
22
CRPS Skill Scores Temperature to Degree Days
.088 -.006
.088 .070
.098 -.027
.098 .082
.086 .090
.086 .053
.088 .093
.088 .085
.108 .097
.108 .066
.106 .081
.106 .085
.110 .092
.110 .060
.074 .078
.074 .049
.109 .038
.109 .090
Skill
.138 .140
.138 .102
.198 .197
.198 .151
High
Moderate
Low
None
.10
.110 .076
.110 .109
.05
T
DD
.01
.104 .095
.104 .074
Cooling
Heating
23
Accumulation Algorithm
DD DD DD Independent (I)
Dependent (D) From Climatology
AB
A
B
F
F
F


2
2
2
B
AB
A
F

F
F

AB
A
B
F
F
lt
F
lt
(I)
(D)
AB
AB
AB
F
F
F

AB
(I)
F

F
F


)
K
K(
AB
F
F
(I)
(D)
(D)
(I)
(D)