DECEMBER 2007 FLOOD EVENT

PART 2: FORECAST VERIFICATION

December 2007 Event Verification

The verification discussion is divided into two sections.In the first section the primary meteorological input, quantitative precipitation forecast (QPF), is evaluated.Evaluation of the hydrologic forecasts follows.

 

Meteorological Verification

This was the first major event where hydrometeorologists at the NWRFC utilized GFE to generate forecasted mean areal precipitation (FMAP) for input to hydrologic models.Hydrometeorolgical forecasters started using GFE generated QPF operationally this season.Each morning the HAS forecaster initializes NWRFC's QPF field using a mix of WFO ISC, HPC, and various model (GFS and NAM) QPF data.The forecaster uses GFE tools to derive a 10 day QPF product.PRISM climatological relationships are utilized to account for the influence of elevation.From the GFE grids QPF amounts are generated for the network of precipitation sites that was used in the NWSRFS model calibration.The FMAP is computed from these points using the calibration MAP weights.

 

Presentation of QPF verification will start off with spatial products generated during the event.This will be followed by basin specific analysis of point forecasts.Figure 1 displays the 24 hour QPF field in GFE for day 1. The derived calibration point precipitation values for that same 24 hour period are shown in figure 2.These can be compared with the image of observed precipitation displayed on the verification web page (see figure 3).


Figure 1 NWRFC GFE display ofday 1 QPF from 12/02/2007 12Z to 12/03/2007 12Z.

Figure 2 NWRFC point QPF plot for day 1 plot from 12/02/2007 12Z to 12/03/2007 12Z

Figure 3 Observed cumulative precipitation from 12/02/2007 12Z to 12/03/2007 12Z.


In fig. 1 there is a 7.45 inch 24 hour day-1 forecast accumulation sample value in the north Oregon Coast Range, along with 7.09 inch accumulation centered over the Washington Olympic Mountains. At the same fig. 1 sample points, the day-2 forecast 24 hour accumulations in fig. 4 are respectively 6.83 and 6.61 inches.The cumulative 2 day precipitation for this two areas are 14.28 and 13.7 inches respectively, which was in the range of observed values.

 

The day 2 GFE QPF grid shown in figure 4 showed that less precipitation was forecasted for Monday. The point QPF product displayed in figure 5 shows that the day 1 GFE QPF grid generated Monday morning (not shown) was an improvement over the forecast two days out.

 

Figure 4 NWRFC GFE display of  day 2 QPF from 12/03/2007 12Z to 12/04/2007 12Z.

 

Figure 5NWRFC point QPF plot for day 1 plot from 12/03/2007 12Z to 12/04/2007 12Z.

Figure 6Observed cumulative precipitation from 12/03/2007 12Z to 12/04/2007 12Z.


Wilson Basin

 

Figures 7 and 8 Comparison of Precipitation Forecasts for the Wilson River Basin.

Figures 9 and 10Scatter Diagrams of Observed Precipitationand NWRFC QPF for the Wilson River Basin.

 

 

Dec 1-3 2007 Event Verification Summary - Wilson Basin Sites

Event Performance Statistics

Event Performance - MAJOR EVENTS

 

 

Event Performance - MAJOR EVENTS

1.00"+ in 6-hour Time Period

 

 

4.00"+ in 24-hour Time Period

 

NWRFC

HPC

PQR

 

 

 

NWRFC

HPC

PQR

POD

0.47

0.77

0.30

 

 

POD

0.50

0.83

0.00

FAR

0.18

0.08

0.10

 

 

FAR

0.40

0.00

1.00

CSI

0.42

0.72

0.29

 

 

CSI

0.38

0.83

0.00

 

 

 

 

 

 

 

 

 

 

Event Performance - MODERATE EVENTS

 

 

Event Performance - MODERATE EVENTS

0.50"+ in 6-hour Time Period

 

 

2.00"+ in 24-hour Time Period

 

NWRFC

HPC

PQR

 

 

 

NWRFC

HPC

PQR

POD

0.90

0.93

0.71

 

 

POD

1.00

0.92

0.75

FAR

0.12

0.11

0.03

 

 

FAR

0.00

0.00

0.00

CSI

0.81

0.83

0.70

 

 

CSI

1.00

0.92

0.75

 

 

 

 

 

 

 

 

 

 

Event Performance - MINOR EVENTS

 

 

Event Performance - MINOR EVENTS

0.25"+ in 6-hour Time Period

 

 

1.00"+ in 24-hour Time Period

 

NWRFC

HPC

PQR

 

 

 

NWRFC

HPC

PQR

POD

0.89

0.93

0.88

 

 

POD

0.89

0.89

0.78

FAR

0.09

0.07

0.09

 

 

FAR

0.00

0.00

0.00

CSI

0.82

0.87

0.81

 

 

CSI

0.89

0.89

0.78

Table 1 Event QPF Verification Statistics for the Wilson River Basin.

 

Dec 1-3 2007 Event Verification Summary - Wilson Basin Sites

Daily Forecast Performance Statistics

Forecast Bias (Mean Algebraic Error)

 

 

Forecast Bias (Mean Algebraic Error)

6-hour Time Period, All Stations

 

 

24-hour Time Period, All Stations

 

NWRFC

HPC

PQR

 

 

 

NWRFC

HPC

PQR

Day 1 Fcst

-0.078

-0.260

-0.236

 

 

Day 1 Fcst

-0.313

-1.038

-0.943

Day 2 Fcst

-0.324

-0.188

-0.476

 

 

Day 2 Fcst

-1.295

-0.750

-1.903

Day 3 Fcst

-0.508

-0.308

-0.610

 

 

Day 3 Fcst

-2.030

-1.232

-2.440

 

 

 

 

 

 

 

 

 

 

All Periods

-0.303

-0.252

-0.441

 

 

All Periods

-1.213

-1.007

-1.762

 

 

 

 

 

 

 

 

 

 

Forecast RMS Error

 

 

Forecast RMS Error

6-hour Time Period, All Stations

 

 

24-hour Time Period, All Stations

 

NWRFC

HPC

PQR

 

 

 

NWRFC

HPC

PQR

Day 1 Fcst

0.682

0.646

0.765

 

 

Day 1 Fcst

1.903

1.455

2.397

Day 2 Fcst

0.858

0.582

0.967

 

 

Day 2 Fcst

1.984

1.229

2.409

Day 3 Fcst

0.979

2.604

1.091

 

 

Day 3 Fcst

2.835

1.745

3.145

 

 

 

 

 

 

 

 

 

 

All Periods

0.848

1.585

0.950

 

 

All Periods

2.280

1.491

2.673

 

 

 

 

 

 

 

 

 

 

Forecast Error (Mean Absolute Error)

 

 

Forecast Error (Mean Absolute Error)

6-hour Time Period, All Stations

 

 

24-hour Time Period, All Stations

 

NWRFC

HPC

PQR

 

 

 

NWRFC

HPC

PQR

Day 1 Fcst

0.405

0.310

0.452

 

 

Day 1 Fcst

1.367

1.075

1.633

Day 2 Fcst

0.592

0.352

0.581

 

 

Day 2 Fcst

1.295

0.790

1.903

Day 3 Fcst

0.617

0.478

0.695

 

 

Day 3 Fcst

2.030

1.232

2.440

 

 

 

 

 

 

 

 

 

 

All Periods

0.538

0.380

0.576

 

 

All Periods

1.564

1.032

1.992

Table 2 Daily QPF Verification Statistics for the Wilson River Basin.


Figures 11 and 12Comparison of Precipitation Forecasts for Jewell, Oregon (JELO3).

 

 

Figures 13 and 14Comparison of Precipitation Forecasts for Lees Camp, Oregon (LEEO3).

 

 

Nehalem Basin

 

Figures 15 and 16Comparison of Precipitation Forecasts for the Nehalem River Basin.

 

Figures 17 and 18 Scatter Diagrams of Observed Precipitation and NWRFC QPF for the Nehalem Basin.


Dec 1-3 2007 Event Verification Summary - Nehalem Basin Sites

Event Performance Statistics

Event Performance - MAJOR EVENTS

 

 

Event Performance - MAJOR EVENTS

1.00"+ in 6-hour Time Period

 

 

4.00"+ in 24-hour Time Period

 

NWRFC

HPC

PQR

 

 

 

NWRFC

HPC

PQR

POD

0.38

0.57

0.26

 

 

POD

0.33

0.42

0.00

FAR

0.16

0.08

0.08

 

 

FAR

0.33

0.00

1.00

CSI

0.36

0.55

0.26

 

 

CSI

0.29

0.42

0.00

 

 

 

 

 

 

 

 

 

 

Event Performance - MODERATE EVENTS

 

 

Event Performance - MODERATE EVENTS

0.50"+ in 6-hour Time Period

 

 

2.00"+ in 24-hour Time Period

 

NWRFC

HPC

PQR

 

 

 

NWRFC

HPC

PQR

POD

0.92

0.93

0.68

 

 

POD

0.94

0.89

0.67

FAR

0.10

0.08

0.02

 

 

FAR

0.00

0.00

0.00

CSI

0.83

0.86

0.67

 

 

CSI

0.94

0.89

0.67

 

 

 

 

 

 

 

 

 

 

Event Performance - MINOR EVENTS

 

 

Event Performance - MINOR EVENTS

0.25"+ in 6-hour Time Period

 

 

1.00"+ in 24-hour Time Period

 

NWRFC

HPC

PQR

 

 

 

NWRFC

HPC

PQR

POD

0.90

0.92

0.89

 

 

POD

0.92

0.92

0.83

FAR

0.08

0.07

0.07

 

 

FAR

0.00

0.00

0.00

CSI

0.84

0.86

0.83

 

 

CSI

0.92

0.92

0.83

Table 3 Event QPF Verification Statistics for the Nehalem River Basin.

 

Dec 1-3 2007 Event Verification Summary - Nehalem Basin Sites

Daily Forecast Performance Statistics

Forecast Bias (Mean Algebraic Error)

 

 

Forecast Bias (Mean Algebraic Error)

6-hour Time Period, All Stations

 

 

24-hour Time Period, All Stations

 

NWRFC

HPC

PQR

 

 

 

NWRFC

HPC

PQR

Day 1 Fcst

-0.112

-0.289

-0.242

 

 

Day 1 Fcst

-0.448

-1.157

-0.968

Day 2 Fcst

-0.321

-0.236

-0.454

 

 

Day 2 Fcst

-1.283

-0.944

-1.816

Day 3 Fcst

-0.475

-0.327

-0.569

 

 

Day 3 Fcst

-1.900

-1.308

-2.278

 

 

 

 

 

 

 

 

 

 

All Periods

-0.303

-0.284

-0.422

 

 

All Periods

-1.210

-1.136

-1.687

 

 

 

 

 

 

 

 

 

 

Forecast RMS Error

 

 

Forecast RMS Error

6-hour Time Period, All Stations

 

 

24-hour Time Period, All Stations

 

NWRFC

HPC

PQR

 

 

 

NWRFC

HPC

PQR

Day 1 Fcst

0.610

0.600

0.687

 

 

Day 1 Fcst

1.724

1.563

2.174

Day 2 Fcst

0.779

0.588

0.860

 

 

Day 2 Fcst

1.846

1.381

2.267

Day 3 Fcst

0.907

0.686

1.016

 

 

Day 3 Fcst

2.587

1.716

2.902

 

 

 

 

 

 

 

 

 

 

All Periods

0.775

0.626

0.865

 

 

All Periods

2.088

1.559

2.469

 

 

 

 

 

 

 

 

 

 

Forecast Error (Mean Absolute Error)

 

 

Forecast Error (Mean Absolute Error)

6-hour Time Period, All Stations

 

 

24-hour Time Period, All Stations

 

NWRFC

HPC

PQR

 

 

 

NWRFC

HPC

PQR

Day 1 Fcst

0.354

0.326

0.404

 

 

Day 1 Fcst

1.177

1.181

1.457

Day 2 Fcst

0.524

0.365

0.537

 

 

Day 2 Fcst

1.283

0.971

1.816

Day 3 Fcst

0.566

0.454

0.645

 

 

Day 3 Fcst

1.900

1.308

2.278

 

 

 

 

 

 

 

 

 

 

All Periods

0.481

0.382

0.529

 

 

All Periods

1.453

1.153

1.850

Table 4 Daily QPF Verification Statistics for the Nehalem River Basin.

 

Figures 19 and 20 Comparison of Precipitation Forecasts for Lees Camp, Oregon (LEEO3).

 

Figures 21 and 22 Comparison of Precipitation Forecasts for Vernonia, Oregon (VRNO3).

 

 

Chehalis Basin

 

Figures 23 and 24Comparison of Precipitation Forecasts for the Chehalis River Basin.

Figures 24 and 25 Scatter Diagrams of Observed Precipitation and NWRFC QPF for the Chehalis Basin.

 

Dec 1-3 2007 Event Verification Summary - Chehalis Basin Sites

Event Performance Statistics

Event Performance - MAJOR EVENTS

 

 

Event Performance - MAJOR EVENTS

1.00"+ in 6-hour Time Period

 

 

4.00"+ in 24-hour Time Period

 

NWRFC

HPC

SEW

 

 

 

NWRFC

HPC

SEW

POD

0.38

0.39

0.56

 

 

POD

0.46

0.38

0.56

FAR

0.34

0.18

0.32

 

 

FAR

0.27

0.00

0.29

CSI

0.32

0.36

0.44

 

 

CSI

0.39

0.38

0.45

 

 

 

 

 

 

 

 

 

 

Event Performance - MODERATE EVENTS

 

 

Event Performance - MODERATE EVENTS

0.50"+ in 6-hour Time Period

 

 

2.00"+ in 24-hour Time Period

 

NWRFC

HPC

SEW

 

 

 

NWRFC

HPC

SEW

POD

0.72

0.89

0.93

 

 

POD

0.59

0.86

1.00

FAR

0.25

0.25

0.22

 

 

FAR

0.14

0.11

0.33

CSI

0.58

0.69

0.74

 

 

CSI

0.54

0.77

0.67

 

 

 

 

 

 

 

 

 

 

Event Performance - MINOR EVENTS

 

 

Event Performance - MINOR EVENTS

0.25"+ in 6-hour Time Period

 

 

1.00"+ in 24-hour Time Period

 

NWRFC

HPC

SEW

 

 

 

NWRFC

HPC

SEW

POD

0.83

0.86

0.95

 

 

POD

0.97

0.96

1.00

FAR

0.15

0.14

0.15

 

 

FAR

0.01

0.03

0.05

CSI

0.72

0.75

0.81

 

 

CSI

0.96

0.94

0.95

Table 5 Event QPF Verification Statistics for the Chehalis River Basin.

 

 

Dec 1-3 2007 Event Verification Summary - Chehalis Basin Sites

Daily Forecast Performance Statistics

Forecast Bias (Mean Algebraic Error)

 

 

Forecast Bias (Mean Algebraic Error)

6-hour Time Period, All Stations

 

 

24-hour Time Period, All Stations

 

NWRFC

HPC

SEW

 

 

 

NWRFC

HPC

SEW

Day 1 Fcst

-0.042

-0.120

0.126

 

 

Day 1 Fcst

-0.169

-0.479

0.502

Day 2 Fcst

-0.119

-0.102

-0.084

 

 

Day 2 Fcst

-0.477

-0.410

-0.336

Day 3 Fcst

-0.189

-0.098

-0.062

 

 

Day 3 Fcst

-0.754

-0.393

-0.248

 

 

 

 

 

 

 

 

 

 

All Periods

-0.117

-0.107

-0.007

 

 

All Periods

-0.467

-0.427

-0.027

 

 

 

 

 

 

 

 

 

 

Forecast RMS Error

 

 

Forecast RMS Error

6-hour Time Period, All Stations

 

 

24-hour Time Period, All Stations

 

NWRFC

HPC

SEW

 

 

 

NWRFC

HPC

SEW

Day 1 Fcst

0.517

0.436

0.654

 

 

Day 1 Fcst

1.249

1.183

2.095

Day 2 Fcst

0.481

0.432

0.573

 

 

Day 2 Fcst

1.282

1.183

1.827

Day 3 Fcst

0.549

0.431

0.589

 

 

Day 3 Fcst

1.572

1.123

1.834

 

 

 

 

 

 

 

 

 

 

All Periods

0.517

0.433

0.606

 

 

All Periods

1.375

1.163

1.923

 

 

 

 

 

 

 

 

 

 

Forecast Error (Mean Absolute Error)

 

 

Forecast Error (Mean Absolute Error)

6-hour Time Period, All Stations

 

 

24-hour Time Period, All Stations

 

NWRFC

HPC

SEW

 

 

 

NWRFC

HPC

SEW

Day 1 Fcst

0.321

0.269

0.422

 

 

Day 1 Fcst

0.873

0.781

1.427

Day 2 Fcst

0.298

0.275

0.342

 

 

Day 2 Fcst

0.827

0.719

1.160

Day 3 Fcst

0.335

0.276

0.356

 

 

Day 3 Fcst

1.007

0.712

1.237

 

 

 

 

 

 

 

 

 

 

All Periods

0.318

0.273

0.373

 

 

All Periods

0.902

0.737

1.274

Table 6 Daily QPF Verification Statistics for the Chehalis River Basin.

 

Figures 26 and 27Comparison of Precipitation Forecasts for Frances, Washington (FRAW1).

 

Figures 27 and 28Comparison of Precipitation Forecasts for Grays Harbor, Washington (CGSW1).

 

 

Skokomish Basin

 

Figures 29 and 30  Comparison of Precipitation Forecasts for the Skokomish River Basin.

 

Figures 31 and 32 Scatter Diagrams of Observed Precipitation and NWRFC QPF for the Skokomish Basin.

Dec 1-3 2007 Event Verification Summary - Skokomish Basin Sites

Event Performance Statistics

Event Performance - MAJOR EVENTS

 

 

Event Performance - MAJOR EVENTS

1.00"+ in 6-hour Time Period

 

 

4.00"+ in 24-hour Time Period

 

NWRFC

HPC

SEW

 

 

 

NWRFC

HPC

SEW

POD

0.57

0.65

1.00

 

 

POD

0.41

0.33

1.00

FAR

0.39

0.22

0.67

 

 

FAR

0.21

0.00

0.25

CSI

0.41

0.55

0.33

 

 

CSI

0.37

0.33

0.75

 

 

 

 

 

 

 

 

 

 

Event Performance - MODERATE EVENTS

 

 

Event Performance - MODERATE EVENTS

0.50"+ in 6-hour Time Period

 

 

2.00"+ in 24-hour Time Period

 

NWRFC

HPC

SEW

 

 

 

NWRFC

HPC

SEW

POD

0.80

0.84

1.00

 

 

POD

0.97

1.00

1.00

FAR

0.14

0.15

0.37

 

 

FAR

0.09

0.08

0.50

CSI

0.71

0.73

0.63

 

 

CSI

0.89

0.92

0.50

 

 

 

 

 

 

 

 

 

 

Event Performance - MINOR EVENTS

 

 

Event Performance - MINOR EVENTS

0.25"+ in 6-hour Time Period

 

 

1.00"+ in 24-hour Time Period

 

NWRFC

HPC

SEW

 

 

 

NWRFC

HPC

SEW

POD

0.90

0.86

0.89

 

 

POD

0.97

0.92

1.00

FAR

0.10

0.03

0.08

 

 

FAR

0.03

0.00

0.00

CSI

0.82

0.84

0.83

 

 

CSI

0.95

0.92

1.00

Table 7 Event QPF Verification Statistics for the Skokomish River Basin.

 

 

 

Dec 1-3 2007 Event Verification Summary - Skokomish Basin Sites

Daily Forecast Performance Statistics

Forecast Bias (Mean Algebraic Error)

 

 

Forecast Bias (Mean Algebraic Error)

6-hour Time Period, All Stations

 

 

24-hour Time Period, All Stations

 

NWRFC

HPC

SEW

 

 

 

NWRFC

HPC

SEW

Day 1 Fcst

-0.178

-0.325

0.416

 

 

Day 1 Fcst

-0.713

-1.300

1.663

Day 2 Fcst

-0.293

-0.286

0.178

 

 

Day 2 Fcst

-1.170

-1.143

0.713

Day 3 Fcst

-0.257

-0.257

0.252

 

 

Day 3 Fcst

-1.028

-1.027

1.007

 

 

 

 

 

 

 

 

 

 

All Periods

-0.243

-0.289

0.282

 

 

All Periods

-0.970

-1.157

1.128

 

 

 

 

 

 

 

 

 

 

Forecast RMS Error

 

 

Forecast RMS Error

6-hour Time Period, All Stations

 

 

24-hour Time Period, All Stations

 

NWRFC

HPC

SEW

 

 

 

NWRFC

HPC

SEW

Day 1 Fcst

0.893

0.876

0.785

 

 

Day 1 Fcst

1.742

2.238

2.336

Day 2 Fcst

0.914

0.808

0.315

 

 

Day 2 Fcst

2.282

2.023

1.058

Day 3 Fcst

0.883

0.784

0.391

 

 

Day 3 Fcst

2.185

1.875

1.235

 

 

 

 

 

 

 

 

 

 

All Periods

0.897

0.824

0.538

 

 

All Periods

2.083

2.051

1.643

 

 

 

 

 

 

 

 

 

 

Forecast Error (Mean Absolute Error)

 

 

Forecast Error (Mean Absolute Error)

6-hour Time Period, All Stations

 

 

24-hour Time Period, All Stations

 

NWRFC

HPC

SEW

 

 

 

NWRFC

HPC

SEW

Day 1 Fcst

0.524

0.472

0.509

 

 

Day 1 Fcst

1.273

1.640

1.663

Day 2 Fcst

0.465

0.436

0.212

 

 

Day 2 Fcst

1.502

1.446

0.713

Day 3 Fcst

0.532

0.417

0.275

 

 

Day 3 Fcst

1.444

1.374

1.007

 

 

 

 

 

 

 

 

 

 

All Periods

0.507

0.441

0.332

 

 

All Periods

1.407

1.486

1.128

Table 8 Daily QPF Verification Statistics for the Skokomish River Basin.

 

Figures 33 and 34Comparison of Precipitation Forecasts for Cushman Dam, Washington (CSHW1).

 

 

Figures 35 and 36Comparison of Precipitation Forecasts for Aberdeen, Washington (ABEW1).

 

Elwha Basin

 

Figures 37 and 38Comparison of Precipitation Forecasts for the Elwha River Basin.

 

 

Figures 39 and 40 Scatter Diagrams of Observed Precipitation and NWRFC QPF for the Elwha River Basin.

 

Dec 1-3 2007 Event Verification Summary - Elwa Basin Sites

Event Performance Statistics

Event Performance - MAJOR EVENTS

 

 

Event Performance - MAJOR EVENTS

1.00"+ in 6-hour Time Period

 

 

4.00"+ in 24-hour Time Period

 

NWRFC

HPC

SEW

 

 

 

NWRFC

HPC

SEW

POD

0.42

0.21

 

 

 

POD

 

 

 

FAR

0.17

0.00

 

 

 

FAR

1.00

 

 

CSI

0.38

0.21

 

 

 

CSI

0.00

 

 

 

 

 

 

 

 

 

 

 

 

Event Performance - MODERATE EVENTS

 

 

Event Performance - MODERATE EVENTS

0.50"+ in 6-hour Time Period

 

 

2.00"+ in 24-hour Time Period

 

NWRFC

HPC

SEW

 

 

 

NWRFC

HPC

SEW

POD

0.74

0.69

 

 

 

POD

0.60

0.60

 

FAR

0.18

0.17

 

 

 

FAR

0.25

0.00

 

CSI

0.63

0.60

 

 

 

CSI

0.50

0.60

 

 

 

 

 

 

 

 

 

 

 

Event Performance - MINOR EVENTS

 

 

Event Performance - MINOR EVENTS

0.25"+ in 6-hour Time Period

 

 

1.00"+ in 24-hour Time Period

 

NWRFC

HPC

SEW

 

 

 

NWRFC

HPC

SEW

POD

0.80

0.82

 

 

 

POD

1.00

1.00

 

FAR

0.10

0.07

 

 

 

FAR

0.00

0.00

 

CSI

0.74

0.77

 

 

 

CSI

1.00

1.00

 

Table 9 Event QPF Verification Statistics for the Elwha River Basin.

 

 

Dec 1-3 2007 Event Verification Summary - Elwa Basin Sites

Daily Forecast Performance Statistics

Forecast Bias (Mean Algebraic Error)

 

 

Forecast Bias (Mean Algebraic Error)

6-hour Time Period, All Stations

 

 

24-hour Time Period, All Stations

 

NWRFC

HPC

SEW

 

 

 

NWRFC

HPC

SEW

Day 1 Fcst

0.004

-0.224

 

 

 

Day 1 Fcst

0.014

-0.897

 

Day 2 Fcst

-0.111

-0.125

 

 

 

Day 2 Fcst

-0.444

-0.500

 

Day 3 Fcst

-0.132

-0.097

 

 

 

Day 3 Fcst

-0.529

-0.388

 

 

 

 

 

 

 

 

 

 

 

All Periods

-0.080

-0.149

 

 

 

All Periods

-0.320

-0.595

 

 

 

 

 

 

 

 

 

 

 

Forecast RMS Error

 

 

Forecast RMS Error

6-hour Time Period, All Stations

 

 

24-hour Time Period, All Stations

 

NWRFC

HPC

SEW

 

 

 

NWRFC

HPC

SEW

Day 1 Fcst

0.417

0.382

 

 

 

Day 1 Fcst

0.688

1.121

 

Day 2 Fcst

0.326

0.269

 

 

 

Day 2 Fcst

0.837

0.677

 

Day 3 Fcst

0.363

0.235

 

 

 

Day 3 Fcst

0.969

0.601

 

 

 

 

 

 

 

 

 

 

 

All Periods

0.370

0.302

 

 

 

All Periods

0.839

0.832

 

 

 

 

 

 

 

 

 

 

 

Forecast Error (Mean Absolute Error)

 

 

Forecast Error (Mean Absolute Error)

6-hour Time Period, All Stations

 

 

24-hour Time Period, All Stations

 

NWRFC

HPC

SEW

 

 

 

NWRFC

HPC

SEW

Day 1 Fcst

0.291

0.250

 

 

 

Day 1 Fcst

0.559

0.897

 

Day 2 Fcst

0.203

0.206

 

 

 

Day 2 Fcst

0.673

0.500

 

Day 3 Fcst

0.253

0.169

 

 

 

Day 3 Fcst

0.798

0.419

 

 

 

 

 

 

 

 

 

 

 

All Periods

0.249

0.208

 

 

 

All Periods

0.677

0.605

 

Table 10 Daily QPF Verification Statistics for the Elwha River Basin.


Figures 40 and 41Comparison of Precipitation Forecasts for Quilcene, Washington (QILW1).

 

Figures 42 and 43Comparison of Precipitation Forecasts for Elwha RS, Washington (ELHW1).

 

 

Flood Forecast Verification

 

Heavy precipitation on December 2 and 3, 2007 caused extensive flooding from the central Oregon Coast up to the Olympic Peninsula of Washington.The storm pushed 34 rivers above flood stage, 6 above major flood stage and 5 forecast locations set new record stages (see tables 1 and 2).For most of these rivers this was the most severe flooding since February of 1996, though the areal extent of this storm was less.A detailed analysis of the river forecasts was performed for the areas that were hardest hit.These areas were the Wilson, Nehalem, Chehalis, Skokomish and Elwha basins.

 

 

 

 

 

Peak

Forecast Point

NWSID

Stage

Time of Peak

Existing Record

Date

 

 

(ft)

 

(ft)

 

Mary's River near Philomath

PHIO3

21.02

12/4 05:00Z

21.07

12/30/2005

Luckiamute River near Suver

SUVO3

29.02

12/4 17:00Z

34.52

12/22/1964

South Yamhill at McMinnville

MCMO3

55.95

12/4 23:45Z

59.33

02/09/1996

Tualatin River near Dilley

DLLO3

19.03

12/4 05:15Z

19.34

12/22/1964

Tualatin River at Farmington

FRMO3

32.66

12/6 02:30Z

37.12

02/09/1996

Johnson Creek at Sycamore

SYCO3

11.92

12/3 18:30Z

14.68

12/22/1964

Clackamas River at Estacada

ESTO3

20.38

12/4 08:00Z

27.57

02/07/1996

Siuslaw River near Mapleton

MPLO3

21.47

12/4 05:30Z

30.21

02/07/1996

Alsea River near Tidewater

TIDO3

21.81

12/4 05:30Z

27.44

12/22/1964

Siletz River at Siletz

SILO3

19.22

12/3 22:30Z

28.62

11/26/2000

Nestucca River near Beaver

BEAO3

18.86

12/3 20:00Z

23.40

11/20/1962

Wilson River near Tillamook

TLMO3

20.45

12/3 19:30Z

22.84

11/07/2006

Nehalem River near Foss

FSSO3

24.44

12/3 19:30Z

29.56

02/08/1996

Willapa River near Willapa

WILW1

26.65

12/4 00:45Z

27.28

12/20/1994

Naselle River near Naselle

NASW1

14.74*

12/3 16:30Z

19.26

03/18/1997

Cowlitz River at Kelso

KELW1

21.6*

12/4 00:15Z

29.50

02/09/1996

Chehalis River near Centralia

CENW1

74.78

12/4 11:30Z

74.31

02/09/1996

Chehalis River near Grand Mound

CGMW1

20.23

12/4 18:00Z

19.98

02/09/1996

Chehalis River near Porter

CRPW1

26.05

12/5 08:00Z

25.22

02/09/1996

Satsop River near Satsop

SATW1

36.63

12/4 03:30Z

38.87

03/19/1997

Puyallup River near Orting

ORTW1

8.35

12/4 02:15Z

11.37

02/08/1996

Nisqually River near National

NISW1

11.29

12/4 01:15Z

12.18

02/08/1996

Deschutes River near Rainier

DSRW1

13.75

12/4 04:00Z

17.01

01/09/1990

Skokomish River near Potlatch

SRPW1

18.12

12/3 22:30Z

17.75

03/19/1997

Elwha River at McDonald Bridge

ELWW1

24.63

12/3 23:00Z

24.20

11/26/1949

Skykomish River near Gold Bar

GLBW1

16.82

12/4 02:00Z

22.49

11/24/1990

Snoqualmie River near Snoqualmie

SQUW1

14.31

12/4 04:30Z

21.55

11/24/1990

Tolt River near Carnation

TOLW1

9.65

12/3 21:45Z

12.92

02/09/1951

Snoqualmie River near Carnation

CRNW1

56.99

12/4 18:00Z

60.70

11/24/1990

Snohomish River near Monroe

MROW1

16.32

12/4 15:30Z

25.30

11/25/1990

Snohomish River at Snohomish

SNAW1

27.87

12/4 22:00Z

30.12

02/10/1951

S.F. Stillaguamish near Granite Falls

GFLW1

14.96

12/4 00:30Z

19.70

02/26/1932

Stillaguamish near Arlington

ARLW1

15.18

12/4 05:39Z

21.06

11/06/2006

Skagit River near Concrete

CONW1

30.60

12/4 13:15Z

42.21

11/21/2003

 

Table 11: December 2007 event peak discharge and peak of record.

 

 


Table 12: Date and time where stage rose above action stages.

 

 

Wilson River

 

Most of the rivers draining Oregon's Coast Range flooded during the early December, 2007 storm.They started to rise early Sunday morning (December 2, 2007), but the rate that the river rose accelerated Sunday night (see figure 44).By 10:00pm Sunday night the Wilson River at Tillamook reached the flood stage of 12 feet (see figure 46).   The river rose passed the major flood stage of 19 feet at 5:30am Monday morning and peak at 20.4 feet around 11:00am that morning.It was not until Tuesday morning that the river fell below flood stage.

 

 

 

 

 

Figure 44 Hydrograph of the Wilson River near Tillamook, Oregon

 

The first forecast issued by the NWRFC, that showed the Wilson River would exceed the flood stage of 12 feet, was issued on Sunday 12/02 at 1631Z.This forecast had two distinct peaks with the highest one being 13.5 feet (see figures 45 and 46).The forecast was updated Sunday evening 12/3 0130Z with a single peak forecast with a maximum height of 15.4 feet.The WFO issued its first flood warning for the Wilson River at 12/3 0239Z.At 12/3 0807Z the RFC updated the forecasted peak stage to 16.7 feet to occur at 12/3 1800Z.The WFO followed by issuing a new flood statement reflecting the new forecast.A 21.5 foot forecast which put the river above major flood (19 feet) was produced at 12/3 1108Z approximately 2 hours before the river exceeded major flood stage.At 12/3 1300Z when the river was at 18.2 feet, the WFO issued a statement that major flooding was forecasted.The final pre-peak forecast was issued at 12/3 1625Z.This forecast put the peak at 22.6 feet to occur at 12/4 0000Z.The river eventually peaked at 20.45 feet at 1930Z on December 3.


Figure 45 Peak Stage Forecast Lead Time for the Wilson River near Tillamook (TLMO3).

 

Figure 46 River Stage Forecasts Issued for the Wilson River near Tillamook.

 

 

Nehalem River

 

Adjacent to the Wilson River, the Nehalem River near Foss showed a similar rise in response to the heavy precipitation (see figure 47).By 12:30am Monday morning the Nehalem River also rose above the flood stage of 14 feet (see figure 49).It continued to rise until 11:30am when it peaked at 24.4 feet, more than 10 feet above flood stage.The flood recession is slow in this basin due to its shape.It is not until Thursday morning that the Nehalem River drops back below flood stage.

 

 

Figure 47 Hydrograph of the Nehalem River near Foss, Oregon.

 

The first forecast that indicated that the Nehalem River would flow over flood stage was issued on 12/1 at 1643Z.The peak stage was forecasted to reach 15 feet (see figures 48 and 49).On the following day at 1631Z the forecasted peak stage was raised to 18.2 feet.This forecast was issued while the river was running at 6.37 feet and precipitation at the Vernonia gage was falling at around 0.2 inches/hours.At 0130Z on 12/3 the forecasted peak stage was raised to 19.9 feet which prompted the issuance of the first flood warning at 0239Z.The forecast was updated at 12/3 0807Z with a higher, earlier peak of 21.5 feet at 12/4 0000Z.The first forecast above major flood was issued on 12/3 1108Z about 8 hours before major flood stage was exceeded.A peak stage of 24.7 feet was forecasted to occur at 12/4 0000Z.At this time the river was running at 17 feet and 4.8 inches of precipitation had fallen at Vernonia at rates approaching 0.5 inches/hour.The flood statement issued at 12/3 1300Z stated that -MINOR FLOODING IS OCCURRING AND MAJOR FLOODING IS FORECAST- with a forecast crest near 25 feet. The final forecast prior to the Nehalem peaking occurred at 12/3 1635Z.This forecast put the peak at 27.7 feet occurring at 12/4 0000Z and 6.9 inches of precipitation had already fallen in Vernonia. The river finally crested at 24.4 feet at 12/3 at 2000Z.

 

 

Figure 48Peak Stage Forecast Lead Time for the Nehalem River near Foss (FSSO3).

 

 

 

Figure 49 River Stage Forecasts Issued for the Nehalem River near Foss.

 

Chehalis River

 

The Chehalis River experienced major flooding and the heaviest impacts to property.The unusual aspect of this event was that the extreme disparity in runoff from different portions of the basin.The Skookumchuck River which is the major tributary draining the northern reaches of the headwaters, didn't even reach flood stage. (Storage of flood waters in the Skookumchuck Reservoir certainly helped attenuate flows but no storage or outflow data is available from this small reservoir to determine its impact on downstream flooding.)Towards the southern headwaters of the basin, the stage of the Chehalis River at the Doty gage exceeded 30 feet before the gage malfunctioned (see figure 50).This stage was 10 feet higher than the previous record stage set on February 8, 1996 and 7 feet above the top of the rating table.Data from this gage proved invaluable in forecasting the record flooding that subsequently occurred downstream.Figure 51 shows the dramatic rise that the residents of Centralia experienced early Monday morning, when the river rose 10 feet in 7 hours after the river was already above flood stage.

 

Figure 50  Hydrograph of the Chehalis River near Doty, Washington.

The first forecast issued, that showed that the Chehalis River at Centralia was going to flood, was issued on Saturday (12/01) at 1752Z.A peak stage just above the flood stage of 65 feet was forecasted to occur at 12/05 0000Z (see figures 52 and 53).Doty was forecasted to peak at 12.2 feet with this model run.On the following day Centralia's peak was increased to 67.4 feet.By Sunday evening (12/03 0240Z) the peak was increased to 68.2 feet. The river was running at 57.9 feet when the forecast was issued.By early Monday morning 7.8 inches of precipitation had fallen during the previous 24 hours causing a rapid rise in the Chehalis River near Doty.This prompted the forecasters to rerunning the model for the Chehalis Basin.The updated forecast issued at 12/03 1218Z put the Doty peak at a near record 20.3 feet and 70.3 feet was forecasted for Centralia.At this time the stage at Centralia was 59.5 feet, still well below flood stage.

 

Figure 51 Hydrograph of the Chehalis River near Centralia, Washington.

 

 

The next model run was made at 12/03 1644Z after the previous day's temperature and precipitation observations were quality controlled and new QPF was issued.At Doty a record stage of 27.5 feet was forecasted.The river was flowing at 26.8 feet when the forecast was issued.The first major flood at Centralia was forecasted at this time.A peak of 73.6 occurring between 12/04 1800Z and 2400Z was forecasted and the Chehalis at Centralia was forecasted to reach flood stage at 12/03 1800Z, earlier than previous forecasts.These forecasts were issued when the river was flowing at 60.9 feet.At 1830Z the Doty gage stops reporting at 30.27 feet, nearly 10 feet above the previous peak.Using the observed peak at Doty the Chehalis forecast group was rerun to generate a new forecast for the downstream Chehalis location.The peak was increased to 74.7 feet to occur at 12/04 1800Z.When the forecast was issued the river was flowing at 61.6 feet and the forecast indicated that the river would rise 6.6 feet in six and a half hours.At 0530Z the river reaches 71.8 feet, a 6 feet rise in two hours. The dramatic rise was primarily the result of water moving down the headwaters of Chehalis River.The Skookumchuck River, the other major tributary upstream of Centralia, did not even reach flood stage during the event.The final pre-peak forecast, issued at 12/04 0615Z, put the peak stage at 76.0 feet occurring at 1200Z.Early Tuesday (12/04) morning at 1130Z the Chehalis River peaked at 74.8 feet.This was about half a foot above the previous record that occurred on February 9, 1996.

 

 

Figure 52Peak Stage Forecast Lead Time for the Chehalis River near Centralia (CENW1).

 

 

Figure 53 River Stage Forecasts Issued for the Chehalis River near Centralia.

 

 

A comparison of the hydrographs from the December 2007 and the February 1996 floods shows a markedly different response in the Chehalis River Basin.The 1996 flood resulted from a rain-on-snow event where the duration of the rain lasted 4 days.The river rose moderately over a 3 day period causing it to crest at 74.3 feet.  In 2007 the river rose quite similarly during the first 30 hours of the event before it lunged above major flood stage in several hours.It can be surmised that the differences were primarily caused by the higher rainfall intensities during the 2007 event, particularly during the Sunday night period.

 

 

Figure 54 Comparison of the Dec 2007 Flood with the Feb 1996 Flood.

 

 

 

Chehalis River near Grand Mound

 

As with the Centralia forecast location, the magnitude of the event at Grand Mound was accurately forecasted more than 24 hours prior to the peak (see figure 56). This was primarily due to the flow observations at the Doty gage.

 

 

Figure 55 Hydrograph of the Chehalis River near Grand Mound, Washington.

 

 

Figure 56 Peak Stage Forecast Lead Time for the Chehalis River near Grand Mound (CGMW1).

 

Figure 57 River Stage Forecasts Issued for the Chehalis River near Grand Mound.

 

 

 

Chehalis River at Porter

 

The Porter gage on the Chehalis River reported some unusual readings during the rising limb of the hydrograph (see figure 58).It appears that it was caused by asynchronous hydrologic response of the river basin upstream of Porter.The initial rise of the hydrograph was cause by the local area response which peaked near the end of 12/03.After the local area response peaked, the drop in flow from the local area was matched by an equivalent increase in flow from higher up in the watershed.After a day of relatively flat flow the upstream peak move downstream pushing the river up another 4 feet.

 

 

 

Figure 58 Hydrograph for the Chehalis River at Porter, Washington.

 

 

Figure 59 Peak Stage Forecast Lead Time for the Chehalis River at Porter (CRPW1).

 

Figure 60 River Stage Forecasts Issued for the Chehalis River at Porter.

 

 

Skokomish River

 

The gage on the Skokomish River near Potlatch set a new record stage on December 3.The new record of 18.12 feet was about a third of a foot higher than the previous peak which was set in 1997.New peak stages have been a regular occurrence at this location because of stream bed aggradation.The stream bed has aggraded several feet during the last half century.As a result the aggradation a new record discharge rate may not have occurred during this event even though there was a record stage.

 

 

Figure 61 Hydrograph for the Skokomish River near Potlatch, Washington.

 

 

Forecasts that put the river above flood stage were issued four days prior to occurrence.This location had the longest lead time of all the forecast locations, which is typical as this river is the one that is most susceptible to flooding.

 

 

Figure 62Peak Stage Forecast Lead Time for the Skokomish River near Potlatch (SRPW1).

 

Figure 63 River Stage Forecasts Issued for the Skokomish River near Potlatch.

 

 

 

Elwha River

 

 

Figure 64 Hydrograph for the Elwha River at McDonald Bridge, Washington.

 

 

 

Figure 65 Peak Stage Forecast Lead Time for the Elwha River at McDonald Bridge (ELWW1).

 

 

Figure 66  River Stage Forecasts Issued for the Elwha River at McDonald Bridge.