2007-2008

Last Year's Forecast

Last Year's Forecast

Last year (August, 2006) we predicted:

• Warmer than average temperatures
• Average to somewhat above-average precipitation
• Good chance of one or more significant snow events in the Portland area (less likely elsewhere in western Oregon)
• Average snowpack in the mountains (possibly higher than average in southern Oregon)
• It could be a really big snow year for Portland.
• For skiers, this winter may not be as good as last year’s (few are!), but it should be decent at least.

We did well. Overall, these predictions came true, at least for western Oregon (much of central and eastern Oregon had a very dry winter).

This Year's Forecast

Below are the factors we used to generate this year's forecast, along with the bottom line for this fall and winter:

• temperatures will be slightly cooler than average in western Oregon and warmer than average east of the Cascades.
• average to somewhat above-average precipitation.
• good chance of low elevation snow in western Oregon.
• above average snowpack in the mountains.

We have also included forecasts from a government agency and three friends, Steve, Teague, and Jordan.

Sea Surface Temperatures

Sea surface temperature (SST) anomalies reported by NOAA/NESDIS on August 27, 2007 show cool anomalies in the eastern Pacific, characteristic of the developing “La Niña” event.

Changes in the tropical Pacific are the single biggest indicators of future weather-climate conditions over the next several seasons. Unfortunately, it is difficult to predict and there are often wide variations in forecasts from different agencies and models. For example:

Climate Prediction Center: La Niña in the fall slowly fading through the winter and spring.
Climate Diagnostics Center (Canonical Correlation Analysis): La Niña developing and continuing.
Climate Diagnostics Center (Linear Inverse Modeling): La Niña developing and continuing.

Meanwhile, the Australian Bureau of Meteorology (BOM) said (August 29):

“Since late July there has been a gradual strengthening of most La Niña indicators. For example, surface temperatures in the eastern equatorial Pacific have approached or just passed La Niña thresholds and there has been further cooling below the surface. Trade Winds have been mostly close to or stronger than normal and cloudiness has been less than average over much of the tropical Pacific.

"However, further strong cooling of the equatorial Pacific Ocean looks unlikely, as computer model guidance suggests ocean temperatures will remain near the La Niña thresholds until the end of the year. In the past, most significant La Niña events were firmly established by winter's end.

"Furthermore, the SOI remains neutral at +2 for the past 30 days, a modest rise from the −4 for the month of July. Continued above average air pressure over Australia has contributed to both the neutral SOI and the below-normal rainfall over much of the country since the start of July. For 2007 to be confirmed as a La Niña year (late-developing by historical standards), the SOI needs to rise to around +7 or higher for several months while the present patterns of Pacific Ocean temperatures, Trade Winds and cloud patterns are maintained."

Our forecast favors a continuation of the current La Niña conditions, but we expect them to level off as a “moderate” La Niña, much as the two CDC models, and the BOM discussion, suggest.

Analog Years

We identified El Niño and La Niña conditions from past years, as well as other physical parameters for which long-term data exist. Using the months January - July, we compared past years with 2007 and identified the closest matches (analogs). These are listed below. The main criteria we used are:

  1. Multi-decadal phase. OCS has identified periods of 20-25 years with generally warm-dry or cool-wet conditions. In each phase, about 75% of all years have been above (wet phase) or below (dry phase) average. The last wet phase was from the late 1940's until the mid-1970's. We believe that we reentered a wet phase in the mid-1990's, making a wetter than average year much more likely than a dry one. Probably the best indicator of these cycles is the Pacific Decadal Oscillation (PDO) parameter. The closest analog years to 2007 are 1945, 1953, and 1990.

  2. The Multivariate ENSO Index (MEI) is based on six main observed variables over the tropical Pacific. These six variables are: sea-level pressure (P), zonal (U) and meridional (V) components of the surface wind, sea surface temperature (S), surface air temperature (A), and total cloudiness fraction of the sky (C). The MEI is computed separately for each of twelve sliding bi-monthly seasons (Dec/Jan, Jan/Feb,...,Nov/Dec). The closest analog years to 2007 are 1952, 1959, 1970, 1978, and 1988.

  3. Other indices. In addition to MEI, OCS examined sea surface temperatures, and the Eastern Pacific (EP), and North Pacific (NP) indices and compared this year's observations with those of previous years. The closest analog years using these indices are 1960, 1973, 1991, and 2001. Another useful index is the Pacific North America (PNA) index; the best analog years are 1953, 1988, and 1998.

  4. Solar cycle. Solar radiation changes are known to have effects on climate, although there is still debate within the climate community regarding the degree and character of those effects. Currently we are approaching a solar minimum period. The closest analog years are 1955, 1965, 1994, and 1996.

  5. Hurricanes. Years ago we noticed a strong correlation between the number of Atlantic hurricanes and Pacific Northwest climate the following winter. A graph of hurricane days vs. Portland winter precipitation appears here. Last year's very busy hurricane season followed by an extreme, active winter in the Northwest is consistent with that pattern. The 2007 hurricane season (so far) has been unremarkable (on the quiet side), suggesting a much less active winter.

Based on a composite of those analyses, the analog years which most closely resemble 2007 (considering the months Jan - July) are 1952, 1953, 1970, and 1988. Using a map composite tool provided by the NOAA-CIRES Climate Diagnostics Center, Boulder Colorado from their website at http://www.cdc.noaa.gov, we created the following temperature and precipitation anomaly maps:

If we simply used the analog years, these would represent our forecast, but by factoring in the effects of La Niña, climate in recent years, and a few other factors, we came up with slightly different forecasts -- see Zone Forecasts below. Also, for the first time in a long time, we are providing monthly forecasts rather than for 3-month periods.

Extreme Weather Events in Analog Years

1952-53: Major flood throughout western Oregon in January. Major windstorm caused significant damage in January.
1953-54: Heavy rainstorm in November in western Oregon. Two strong windstorms in December. Three major wind-rain-snowstorms in January.
1970-71: Major windstorm in March affected entire state. Wet December with low elevation snow. Heavy snow in January (11" in Salem, 9" in Portland).
1988-89: Heavy rain with localized flooding in November. Extremely dry month of December. Record low temperatures and heavy snow fall in February with five deaths. Flooding in central Oregon in March. Major windstorm in March killed three people.

Zone Forecasts

Below are our monthly forecasts for precipitation, temperature, and snow for the months October-March. We have created separate forecasts for Oregon regions west and east of the Cascades. For western Oregon, we have included predictions of the likelihood of low elevation snow.

The numbers in the table represent our estimate of the "percentile" probabilities for each month. For example, a value of 50 would represent an equal chance of above- or below-average (or normal) for that month; values higher than 50 suggest higher probabilities of wetter or warmer than average, while values below 50 indicate increasing chances of drier or cooler. The more extreme the value, the more likely that extreme conditions will occur in that month.

Overall we are predicting western Oregon conditions to be somewhat cooler and wetter than average, with a good chance of at least one low elevation snow event. Central and eastern Oregon are expected to have warmer than average temperatures and average precipitation.

After a slow start, we expect mountain snows to accumulate quickly in November and remain mostly good throughout the winter. January and February mountain snows are expected to be significant. All in all, we anticipate a very good ski season!

Other Forecasts

• Climate Prediction Center

CPC's forecasts include monthly and seasonal (3 months) predictions for precipitation and temperature anomalies. For the upcoming winter they are as follows:

• Steve Pierce's Forecast

• Teague Rakoz's Forecast

• Jordan Brice