La Niña Impacts in the Pacific Northwest
Nathan Mantua
Joint Institute for the Study of the Atmosphere and Oceans
University of Washington
Seattle, WA. USA
Email: mantua@atmos.washington.edu
Introduction
Does knowledge that La Niña conditions will exist over the course of any given period of time offer useful insight into likely climate anomalies for the Pacific Northwest region? If so, should we expect that public and private agencies, as well as individuals, will be able to use La Niña-related climate forecasts to either mitigate expected losses and/or exploit expected opportunities?
In this brief article, empirical relationships between the tropical La Niña and some features of Pacific Northwest climate are documented and discussed. Statistical analyses identify relatively clear and consistent Pacific Northwest climate anomalies in association with the 10 strongest La Niña episodes of the past century. Other studies have found similar strong and reliable teleconnections between La Niña and Pacific Northwest climate. Given the present-day capabilities in monitoring and predicting La Niña (and El Niño) variability at seasonal to interannual time scales, there appears to be great potential for predicting aspects of Pacific Northwest climate during periods with strong La Niña (and El Niño) conditions.
With anomalous climate conditions come potential impacts on climate sensitive sectors and activities. La Niña related impacts in the Pacific Northwest are discussed in Section 3. A general discussion of the use and utility of climate information appears in Section 4.
Pacific Northwest Climate Anomalies Associated with La Niña
Generally speaking, periods with strong La Niña conditions have been associated with anomalously cool, wet climate conditions from October through March in the Pacific Northwest. Conversely, moderate La Niña conditions have not been associated with strong and consistent climate anomalies in the Pacific Northwest.
TABLE 1. Categorization of mean October-March NINO3.4 values in terms of non-ENSO, and strong or moderate La Niña years. The criteria used in this analysis identifies as a strong La Niña episode those October-March periods for which the average NINO3.4 SST index is more negative than -1.2 standard deviations. Moderate La Niña episodes are identified as those October-March periods for which NINO3.4 is between -0.5 and -1.2 standard deviations. The following table lists moderate and strong La Niña years, according to these criteria
1955-56, 1970-71, 1973-74, 1975-76, 1988-89 | |
| Moderate La Niña Years | 1897-98, 1898-99, 1903-04, 1906-07, 1908-09, 1910-11, 1917-18, 1920-21, 1924-25, 1938-39, 1943-44, 1944-45, 1950-51, 1954-55, 1964-65, 1967-68, 1971-72, 1974-75, 1983-84, 1984-85, 1985-86, 1995-96 |
Especially strong changes in the observed frequency distribution of October-to-March precipitation are evident for western Washington state (US Climate Division Data zones 1-5, available via the internet at: www.ncdc.noaa.gov/onlineprod/drought/main.html ). Estimated frequency distributions for observed "strong" and "moderate" La Niña years (Table 1), along with "non-ENSO" years, are plotted in Figure 1. There is a clear shift toward higher than average frequencies of anomalously heavy precipitation for the strong La Niña years, while there is little distinction between the frequency distributions for moderate La Niña and non-ENSO years. October-to-March surface temperature frequency distributions for western Washington state are shifted toward cooler than average values during strong La Niña years, while those for moderate La Niña years are centered on the long-term average (Figure 2). These changes in the surface temperature and precipitation distributions are generally true for the entire Pacific Northwest region (Oregon, Idaho, Washington, and southwestern Canada). The combination of increased frequencies of anomalously cool, wet weather in the region is also evident in regional streamflow and snowpack data. For instance, the frequency distribution for annual water year (October-September) streamflows on the Columbia River show a clear shift towards anomalously high values during strong La Niña years, with sharply increased (decreased) frequencies of extremely high (low) flows, relative to all other years (Figure 3, top panel). Composite hydrographs show that strong La Niña year monthly mean Columbia River flows tend to be especially anomalous in the late spring and early summer months as the regional snowpack accumulated in the fall and winter months is melted (Figure 3, bottom panel).
| FIGURE 1. Estimated Western Washington October-to-March precipitation frequency distributions for non-ENSO (dotted) versus strong (solid) and moderate (dash-dot) La Niña years. Observed anomalies for 1989 and 1996 are plotted and labeled by the star and dot symbols, respectively. |  |
| FIGURE 2. Estimated Western Washington October-to-March temperature frequency distributions for non-ENSO (dotted) versus strong (solid) and moderate (dash-dot) La Niña years. Observed anomalies for 1989 and 1996 are plotted and labeled by the star and dot symbols, respectively. |  |
| FIGURE 3. Strong and moderate La Niña year streamflow anomalies. In the top panel are estimated frequency distributions for Columbia River water year (October-September) streamflows. In the bottom panel are composite naturalized hydrographs for the Columbia River at the Dalles, Oregon. The composite non-ENSO year hydrograph is shown with the gray bars. |  |
Composite La Niña Columbia River Flows
Additional research has found that changes in the frequency distributions for mean precipitation and streamflow are accompanied by similar shifts in daily extremes for these two variables. In the Pacific Northwest, extremely high rainfall and streamflow events (i.e., floods) have occurred at anomalously high frequencies during strong La Niña years relative to all other years in the historical record (not shown).
La Niña-Related Climate Impacts in the Pacific Northwest
Changes in the Pacific Northwest's hydrology, like those identified in Figure 3, produce a variety of impacts on a wide range of natural resource sectors. For instance, the strong tendency for above average water-year streamflows leads to a reduced likelihood for conflicts over water allocation between competing stakeholders in the Columbia River Basin. Likewise, an increased probability for an abundant water supply suggests a similar probability shift toward relatively good in-stream habitat for juvenile anadromous fish (salmon) during the typically low-flow late summer and fall months. On the other hand, the increased risk of extremely high October-to-March streamflow events leads to an increased risk for property damage (flooding) and relatively poor survival of incubating salmon eggs (via scouring of gravel nests).
The tendency for anomalously cool October-to-March weather is also expressed as an increased frequency of low elevation snowfall events in the Pacific Northwest. Such storms often produce large financial losses in urban areas of the Puget Sound Lowlands. At the same time, a tendency for abundant snowpack is welcome news for ski enthusiasts and water managers.
Finally, the tendency for anomalously cool October-to-March surface temperatures is also true for coastal ocean temperatures. Anomalous cooling of the coastal ocean (along with a suite of related changes to the upper ocean environment) tends to provide favorable conditions for anomalously high biological productivity for a number of species, including top level predators like sea birds, marine mammals and Pacific salmon.
Discussion
Assuming there is a demonstrated capability to predict climate anomalies at seasonal to interannual time scales, can we assume that society as a whole will benefit from this technology? The societal response to the climate anomalies attributed to the 1997/98 El Niño episode suggests that the use of climate information may be as unpredictable and mysterious as the climate system itself.
During the 1997/98 El Niño the climate forecasting and societal impacts communities were faced with a range of challenges. These included purely scientific questions of climate forecast accuracy, to the flow of conflicting climate information from multiple sources, and then to societal constraints imposed by such things as institutional barriers to the use of climate information. Generally speaking, the problems in utilizing El Niño-based climate forecasts are identical to those that will limit the use of La Niña-related climate information.
Studies conducted over the past 3 years by the Climate Impacts Group at the University of Washington find that, in spite of the difficult hurdles listed above, there are relatively fast and inexpensive ways to improve the utility and value of climate forecasts. One route is to extend climate forecasts into impacts forecasts. In this case, impacts are viewed as the consequences of an increased probability of physical climate anomalies such as an unusually heavy snowpack. The impacts outlook would take this information and extend it into a prediction for something like an increased probability for an outstanding ski season. The important distinction here is that the climate information is translated into a language that is understandable to a wider audience than that of the climate prediction community.
A second and complimentary route is to maintain an open dialogue with the so-called user community. In many cases, the climate research and prediction community generates many products that are largely unintelligible and therefore useless to both the general public and decision makers in climate-sensitive sectors. Increased utility of climate information appears to be closely related to value-added information, translation, and directed application to specific issues. Presently, there does not appear to be a lack of climate information, but more likely there is more data and analyses than the non-specialist can digest. It is hoped that concerted efforts to foster collaboration between the climate research, climate impacts and user communities will quickly extend the usability and use of climate information