Modeling, Simulating, and Forecasting Subseasonal Atmospheric Variability I: Diagnostics
Presiding: D Waliser, California Institute of Technology; K Weickmann, National Oceanic and Atmospheric Administration
A31A-01 INVITED 08:30h
Dynamics of the most predictable patterns in week two forecasts
The phenomena that yield skill in the second week of a forecast are generally large scale and low frequency, and hence there may be only a few independent samples of these events each season. In addition, the predictable signal may be small compared to the uncertainty in a single forecast, so ensembles may be needed to extract that signal. Therefore, quantifying the nature of the predictable signal in week two therefore requires a large sample of ensemble forecasts with a fixed model, spanning many years. Using the CDC MRF reforecast dataset (a 25-year dataset of ensemble forecasts with a fixed model), we will attempt to shed light on a few basic questions such as "How much predictive skill is there in week 2?" and "Where does that skill come from?". A canonical correlation analysis is performed to isolate the most predictable patterns in week two for Northern Hemisphere winter. The three most predictable patterns are very similar to the "Tropical/Northern Hemisphere" (TNH), "Pacific/North American" (PNA) and "North Atlantic Oscillation" (NAO) teleconnection patterns identified in previous studies of low-frequency variability. Regression analyses are used to elucidate the relevant mechanisms responsible for the remarkable predictability of these patterns, including the role of tropical convective forcing, large-scale eddy-mean flow interactions, and synoptic-scale transient eddy feedbacks.
A31A-02 INVITED 08:45h
Extreme Winter Precipitation Events in the Western United States: The impact of ENSO and the Madden-Julian Oscillation
The west coast of the United States occasionally experiences intense winter storms that account for a major fraction of the total seasonal rain(snow)fall. In some cases, it is not a single storm, but a series of storms, that batter the west coast in a matter of few weeks. These storms, unfortunately, are often associated with flooding, mudslides and other disasters that can lead to extensive property damage and even loss of life. In this talk, I will review our current understanding of the nature of these storms and the extent to which their occurrence is impacted by El Nino/Southern Oscillation and the Madden Julian Oscillation. The results are based on 50 years of precipitation observations, NCEP/NCAR reanalyses, and idealized experiments with a global atmospheric general circulation model.
A31A-03 09:00h
Coherent Life Cycle of Intraseasonal Tropical Convection and Extratropical Circulation during El Nino and La Nina years
Coherent life cycle of intraseasonal tropical convection and extratropical circulation has been studied with the NCEP/NCAR reanalysis and NOAA satellite data during boreal winters from 1979 to 2003. From the EOF analysis, eastward propagating life cycle of tropical and extratropical circulation anomalies are obtained using the composites of 1st and 2nd principal components using OLR. Extratropical circulation anomalies are found to be closely connected with tropical convection cells in 20-90 day life cycle. During El Nino, intraseasonal tropical convection intensifies over the eastern Pacific and during La Nina, it is strengthened to a little northward over the western Pacific. From these intraseasonal tropical forcings, mid- to high-latitude wave propagation activity seemed to show distinct contrast features during ENSO years. However, regions of north Atlantic and eastern part of North America showed no significant reactions from the tropical forcings. Wave-eddy feedbacks and the eddy energy over the extratropics are calculated to find the possible other reasons for this differences.
A31A-04 09:15h
The Synoptics of Subseasonal Forecasting: Lessons From Weather-Climate Monitoring
The synoptic events that encompass the subseasonal time band are generally vague and little studied. Variability beyond an individual storm involves a rapidly increasing spatial domain and the interaction of multiple time scales. The events range from the fast local interactions of baroclinic wavetrains and tropical convective flare-ups to the medium time scale interactions of the MJO, teleconnection patterns and surface soil moisture/snow cover to the slow times scales of seasonal SST changes including ENSO. The MJO occupies a unique niche in the time band as it organizes tropical convection at 30-60 day periods, influences mid-latitudes, and can affect both higher and lower frequency variability. Fast dynamical processes (e.g., baroclinic life cycles, wave-mean flow interaction) dominate the variability slowly giving way to MJO time scales (e.g., tropical heating) and then coupled ocean atmosphere dynamics. When boundary forcing is weak seasonal anomalies are often the residual of large amplitude subseasonal events. Since northern Fall 2003 an experimental MJO website has been online (www.cdc.noaa.gov/MJO) gathering and posting subseasonal predictions from a variety of models. Weeks 1 and 2 and day 11 are emphasized on the site although predictions to 30-45 days are being planned. A subseasonal synoptic model has been used to help monitor and analyze synoptic behaviors as part of a ~monthly weather-climate discussion, also available online. Selected events will be used to illustrate the forecasting problem from a synoptic perspective. These will be discussed in the context of signal to noise, linear versus nonlinear and coupled versus uncoupled variability.
A31A-05 09:30h
The Impact of Easterly Waves and the MJO on Moisture Surges over the Gulf of California
In this study we examine the links between moisture surges over the Gulf of California, tropical easterly waves, and tropical heating on time scales of the Madden Julian Oscillation (MJO). The results are based on the NASA TRMM reanalysis and the NCEP North America Regional Reanalysis. Composite analyses are used to show the typical evolution of the circulation and precipitation fields prior to, during, and after major surge events. We show that both the MJO and easterly waves play an important role in modulating the strength and onset of the moisture surges over the Gulf of California. We also examine the ability of an atmospheric general circulation model to simulate gulf surges, and the observed links to the MJO and easterly waves.
A31A-06 09:45h
A Quantitative Approach to Flash Flood Prediction in Southern Utah
Flash flood monitoring and prediction is considered to be a critical part of National Weather Service (NWS) severe weather operations in the semi-arid western United States. The complex terrain and steep slopes in this area, combined with impervious rock and soils, can induce flash flooding with relatively light rainfall. This reduces the value of using the more common conceptual flash flood models developed for the central and eastern United States. Thus, forecasters at the NWS Weather Forecast Office in Salt Lake City, Utah, have relied on a locally developed conceptual model to predict the likelihood of flash flooding on a given day. Until this study, common practice was to assume that humid and unstable air combined with low wind speeds in the lower troposphere would yield rainfall conductive to flash flooding. A new approach to flash flood prediction, exploring the connection between atmospheric variables and flash flood reports, will increase situational awareness and provide forecasters with quantitative flash flood guidance. A record of historical flash floods in southern Utah was compiled to determine the frequency of events from 1959 to 2003. A complete data set, consisting of both historical flash flooding days and non-event days, was assembled. A trial of the 2003 three-month flash flood season assessed which variables and which dataset to use in studying the eight flash flood seasons from 1996 to 2003; the trial concluded that the best source of atmospheric data was a set of soundings from Flagstaff, Arizona, a location close to and generally upstream of southern Utah. Neural networks were used to determine the relationship between the atmospheric state and a particular day's flash flood severity. The final neural network used six input variables and a discretized output variable. Precipitable water, low-level relative humidity, convective available potential energy, the 500hPa height change between 12Z and 0Z the following day, and the previous day's flash flood severity were found to be the important determinants of flash flooding in southern Utah. Data collected throughout the 2004 flash flood season was used to verify the accuracy of the above-mentioned flash flood prediction algorithm.