|
|
|
|
Featured Researcher
Philip Mote
Philip Mote directs OCCRI, co-leads CIRC, also co-leads the Northwest Climate Science Center, has been a contributing author for Intergovernmental Panel on Climate Change's Fourth and Fifth Assessment Report and the Third National Assessment, and yet he still finds time to do original research, teach classes at Oregon State University, be a father to three children, participate on a rowing team, and, last but certainly not least, write for this newsletter. This month’s newsletter features not one, but three articles written by Mote.
|
These 2,000-year-old petrified stumps appear from time to time on the beach in Neskowin as a result of significant erosion in the area. (Harold Zald)
Sea-Level Rise
Past Observations May Greatly Underestimate Future Damage
It is no secret that we get big storms and big waves in the Pacific Northwest. In fact, some are so big that “storm watching” is even considered an important recreational activity by those who keep track of coastal tourism. But for property owners, these storms are serious business. For example, in the winter of 1997 - 1998, El Niño brought exceptionally large storms and wave heights that flooded some homes and irreparably damaged others by eroding the ground beneath them.
Relying only on the observational record may significantly underestimate what areas are at risk along the coastline. Two recent papers by CIRC researchers seek to quantify the likelihood that storm surges and extreme waves will change in the future. One paper found that the 100-year event of extreme water levels (an event that has a 1 percent chance of occurring in a given year) could be nearly 3 feet higher and cause 30 percent more coastal flooding than previously estimated. The other paper found that the number of homes and businesses exposed during a 100-year event under existing land-use policies were likely to double by the 2050s.
Although sea-level rise refers to a slow, inexorable change, actual damage is caused by the highest water level, which may last only a few seconds. That’s because many processes determine water level at any given moment. These processes operate on time scales ranging from decades to seconds, and include sea-level rise, seasonal changes, storm surges, tides, and the familiar rhythmic wind-driven waves.
In order to tease apart, understand, and predict each of these processes separately and together, CIRC researchers Katy Serafin and Peter Ruggiero use a framework they call “total water levels.” Until now, calculating the likelihood of these factors occurring at their extremes during the same point in time — a worst of the worst-case scenario — was based on past experience — that is, by looking over the observational record and assessing how often extreme water levels occurred in the recent or distant past. The limitation of this approach is that these estimates are based on what has occurred, not necessarily what could occur based on our understanding of what is physically possible when all the processes line up the wrong way.
To overcome this problem, Serafin and Ruggiero developed a statistical model to represent the processes driving total water levels and ran it over and over, in effect recreating the past thousands of times to develop different combinations of total water levels and more robust estimates for how likely the extreme combinations of the various water-level components occurred. They applied the model to Tillamook County, on the northern Oregon coast, as part of a CIRC project.
In another recent paper, this one by Baron, Ruggiero and colleagues, a similar approach was used to estimate coastal erosion hazards in Tillamook County by evaluating 2,000 different combinations of future sea-level rise, changes in wave heights, El Niño events, and shoreline types. The study found that changes in wave heights had the most significant influences on total water levels, but that sea-level rise had the most impact on shoreline erosion processes.
Much of the work from these two papers is being used in the Tillamook County Coastal Futures Project, in which Ruggiero, Serafin, and other CIRC researchers are working with coastal stakeholders and policymakers to evaluate the effects of alternative land-use policies on community adaptation to climate change and extreme events over the coming decades.
Serafin, Katherine A. and Peter Ruggiero (2014) Simulating extreme total water levels using a time-dependent, extreme value approach, Journal of Geophysical Research, 119, 9 DOI: 10.1002/2014JC010093
Baron, Heather M., Peter Ruggiero, Nathan J. Wood, Erica L. Harris, Jonathan Allan, Paul D. Komar, Patrick Corcoran (2014) Incorporating climate change and morphological uncertainty into coastal change hazard assessments Natural Hazards DOI 10.1007/s11069-014-14
Ocean Systems
Coastal Upwelling Winds Have Intensified on the Pacific Coast
Summer upwelling winds along the U.S. Pacific Coast have intensified since the 1940s, a recent data analysis shows. This trend has implications for marine life and fisheries, as well as for hypoxia (dead zones) and ocean acidification.
In a “study of studies” (also known as a “meta-analysis”), a team of scientists reviewed more than 20 scientific papers and nearly 190 previous (though not necessarily independent) analyses of wind trends for the world’s major upwelling systems. They found a “preponderance” of studies showing that summer upwelling winds that blow in the southerly direction have intensified since the 1940s. They also noted that summertime winds off the Pacific Northwest got stronger toward the south. The start date of the studies did not change the final results.
Coastal upwelling systems, such as the California Current (the name for the ocean current off the west coast of the US), provide 20 percent of the world’s fish harvest, yet they cover only 2 percent of total ocean area. The upwelling of deeper, nutrient-rich waters is what makes these systems so productive, and it is the surface winds that drive the upwelling. Long-term changes in the winds, therefore, may have a profound physical and biological effect on these systems. While increases in upwelling can benefit marine life, very large increases may also be disruptive by bringing up acidic or hypoxic waters, or moving plankton off the continental shelf.
The role of anthropogenic forcing on upwelling winds is not clear. In their meta-analysis, the researchers were not able to separate relative contributions from natural variability and anthropogenic forcing.
Sydeman, W.J., M. Garcia-Reyes, D. S. Schoeman, R. R. Rykaczewski, S. A. Thompson, B. A. Black, and S. J. Bograd (2014) Climate change and wind intensification in coastal upwelling ecosystems, Science, 345, 77-80, doi: 10.1126/science.1251635.
|
|
In December 2007, the Chehalis River flooded Interstate 5 in Centralia, closing the freeway and impacting north-south travel and commerce for several days. NOAA
Flooding
Mild River Basins More Prone to
Flooding as Region Warms
As the Northwest warms, most river basins will become more prone to flooding, while others will remain relatively unchanged, a study from the University of Washington suggests. Flood risks appear to be linked to winter temperatures (which are largely determined by elevation).
Lower-elevation, warmer-winter basins in Oregon and Washington could tend to flood more as regional temperatures rise, according to the UW team led by Eric Salathé. By running a sophisticated hydrologic model (Variable Infiltration Capacity) along with a regional climate model for both past and future climates, they found higher incidences of 100-year floods in the future for all of the warmer, lower basins. (A 100-year flood is so-called because it has only a 1 percent chance of occurring in any given year.)
Most, but not all, of the cooler basins, also show increases. The coldest basins (those with a mean winter temperature less than -7 degrees Celsius or 19 degrees Fahrenheit) are split down the middle: Some show more flooding, some about the same, some less.
Compared with other studies using regionally averaged results, the recent UW study highlights the regional texture offered by a regional climate model: specifically the winter-temperature dependence of flood risk.
The study’s main conclusions are shown in the figure below, in which the y-axis is the ratio of future (2050s) to past magnitude of 100-year floods, where 1 means no change, and 2 means a doubling. Colors indicate the month when the 100-year flood typically occurs: warm basins in the fall, coldest basins in summer.
Our Editorial Opinion: This is a valuable study and more work is needed, particularly to quantify the range of possibilities. In 2011 a team led by Francina Dominguez from the University of Arizona studied extreme precipitation using data from the North American Regional Climate Change Assessment Program (NARCCAP). The NARCCAP regional models had changes in extreme daily precipitation for the Northwest ranging from negative 5 percent to 30 percent, with an average of about +10%. While those were regionally averaged results, and for precipitation not streamflow, they show that the choice of regional model can strongly affect results. Repeating the UW study with a few different combinations of regional and global climate models would help elucidate how robust these results are.
Salathé, Eric P. Jr., Alan F. Hamlet, Clifford F. Mass, Se-Yeun Lee, Matt Stumbaugh, and Richard Steed (2014). Estimates of Twenty-First-Century Flood Risk in the Pacific Northwest Based on Regional Climate Model Simulations, J. Hydrometeor, 15, 1881–1899. doi: http://dx.doi.org/10.1175/JHM-D-13-0137.1.
|
The ratio of the projected future to historical 100-yr flood as a function of winter (DJF) mean temperature for the (top) bias-corrected ECHAM5-WRF-VIC simulations for the 2050s. Figure from Salathé et al., © American Meteorological Society, used by permission.
Estimating Streamflow
Finding the Best Indices of Drought in the Northwest
Sometimes a simple approach works pretty well. That’s what CIRC PI John Abatzoglou at University of Idaho and his colleagues found when they held a scoring contest to gauge the streamflow estimating power of four drought indices in the Pacific Northwest — something that, of course, can be done at much more expense and time by running a sophisticated hydrologic model. The four indices they chose were: (1) standardized precipitation index; (2) standardized precipitation-evapotranspiration index; (3) Palmer Drought Severity Index; and (4) a water-balance runoff model.
The indices were tested on 21 unregulated Pacific Northwest river basins to determine which ones best represented year-to-year streamflow variability. The indices — all of which used weather data to produce a single number for each year — compared year-to-year variations in drought estimates with streamflow in basins ranging from warm to very cold. Scores were based on the percentage of year-to-year variability in October through September (“water year”) streamflow that could be explained by the contestant.
In most basins, precipitation-evapotranspiration and water-balance runoff were the best approaches, with the precipitation index coming in a distant third. The Palmer Drought Severity Index was nowhere near the frontrunners. Its poor showing, the researchers surmised, could be its failure to distinguish between rainfall and snowfall, and because it is designed for estimating soil moisture, not runoff. The water-balance index did best in mild coastal basins, where it often scored better than 90 percent. The toughest basin to model was Middle Fork Rock Creek in Montana, where the highest score (by the precipitation-evapotranspiration index) was 56 percent.
The Abatzoglou team concluded, among other things, that even these simple approaches depend on having enough measurements of hydrologic variables in the mountains. Operational forecasting of streamflow is much more sophisticated but also relies on mountain observations.
Abatzoglou, John T., Renaud Barbero, Jacob W. Wolf, and Zachary A. Holden (2014) Tracking Interannual Streamflow Variability with Drought Indices in the U.S. Pacific Northwest, J. Hydrometeor, 15, 1900–1912. doi: http://dx.doi.org/10.1175/JHM-D-13-0167.1
|
Moisture flux in kg kg m-1s-1 for December 3, 2007, when a damaging flood occurred in southwestern Washington and northwest Oregon (see photo at the top). Figure from Payne and Gudrun Magnusdottir, © American Meteorological Society, used by permission.
Atmospheric Rivers
Wettest Storms Linked to Planetary-Scale Waves
Some of the wettest storms in the Northwest are “atmospheric rivers.” This is a catchy name for a phenomenon in which an elongated band of warm moist air intersects with the Pacific coast of North America. Researchers long have known that atmospheric river events are linked to their subtropical moisture source. But a new look at climate data shows additional linkages to planetary-scale atmospheric waves in northern latitudes. In a recent analysis, researchers Ashley Payne and Gudrun Magnusdottir found that the number of atmospheric river events peaks in November, decreases steadily until January and then drops rapidly until March. The most common northern latitude for these events is 45 degrees, which falls between Tillamook and Newport on the Oregon coast.
Surprisingly, the researchers also found that the El Nino Southern Oscillation (ENSO) affected atmospheric river events in an unexpected way. Because seasonal precipitation in the Pacific Northwest tends to be higher during La Niña periods than during El Niño periods, one might expect more atmospheric rivers to occur during La Nina. But this study showed the opposite, with more atmospheric rivers occurring during El Niño. To further complicate the picture, most such events occurred during ENSO’s neutral phase (that is, years when ENSO is neither in the El Niño nor La Niña phase).
By tracking large atmospheric river events backward across the Pacific, the researchers linked those events to an atmospheric dynamics phenomenon known as “Rossby wave breaking” (Rossby waves are the large-scale wavy patterns that lead to regular weather systems, typically at latitudes north of 40 degrees). Rossby wave breaking also influences atmospheric rivers, even though they extend into much lower latitudes.
For their study, the researchers used MERRA, a research-quality “reanalysis” dataset to study the patterns of atmospheric rivers. MERRA is a climate model into which atmospheric observations are fed every day. It solves all the same equations as a free-running climate model. Atmospheric rivers are often studied using simple 2-D metrics like moisture flux at a certain level (usually 700 millibars, or roughly 10,000 feet). The use of MERRA allows a more physically complete way to characterize atmospheric rivers, since it’s a fully 3-D dataset. This allowed Payne and Magnusdottir to add up the moisture at all levels to define an atmospheric river.
The study focused on the biggest moisture flux events (95th percentile and up) and especially on a handful of events that exceed both the 95th percentile for moisture flux and the 95th percentile for precipitation.
Payne, Ashley E. and Gudrun Magnusdottir (2014) Dynamics of Landfalling Atmospheric Rivers over the North Pacific in 30 Years of MERRA Reanalysis. J. Climate, 27, 7133–7150. doi: http://dx.doi.org/10.1175/JCLI-D-14-00034.1
|
Moderate Resolution Imaging Spectroradiometer (MODIS) on NASA’s Aqua satellite acquired this natural-color view of a massive bloom of phytoplankton off the coast of Oregon and Washington, common during the summer due to the coastal upwelling process. July 31, 2014. NASA
|
|
The Climate CIRCulator is brought to you by The Pacific Northwest Climate Impacts Research Consortium (CIRC). CIRC delivers science, information, and tools to decision makers responsible for the management of resources and services in a changing climate. Our team consists of experts from Oregon State University, the University of Oregon, the University of Idaho, and the University of Washington. CIRC is funded by the National Oceanic and Atmospheric Administration (NOAA) and housed in the Oregon Climate Change Research Institute (OCCRI) at Oregon State University. The OCCRI brochure can be downloaded here.
|
|
|
|
|
|
