Guidance-on-Guidance for Bifurcations in Seven-day Intensity and Intensity Spread Predictions of Western Pacific Tropical Cyclones

Our Weighted-Analog Intensity technique for western North Pacific (now called WAIP) tropical cyclone intensity and intensity spread guidance has been extended from five days to seven days. Tests with an independent sample from the 2010-2014 seasons demonstrate that the intensity mean absolute errors and the correlation coefficients of the WAIP forecast intensities with the verifying intensities essentially remain constant in the five-day to seven-day forecast interval. After calibration of the raw intensity spreads among the 10 historical analogs each 12 hours, the uncertainty estimates about the WAIP intensity forecasts also do not increase during the five-day to seven-day forecast intervals. As Tsai and Elsberry (2014, Asia-Pacific J. Atmos. Sci.) found for the five-day WAIP version, bifurcation or bi-modal tropical cyclone intensity forecasts arise in the seven-day WAIP due to uncertainty in the timing of formation, timing and magnitude of Rapid Intensification (RI) periods, or track forecast uncertainty leading to landfall or non-landfall. About 19% of the overall sample of five-day forecasts during the 2010-2012 met the criteria for a substantial intensity bifurcation situation. Using a hierarchical clustering technique, two clusters of the 10 best historical analogs were defined and separate WAIP forecasts and intensity spreads were calculated for the two clusters. If an always perfect selection of the correct cluster WAIP forecast of each bifurcation situation was made, a substantial improvement in the intensity mean absolute errors was achieved relative to the original WAIP forecasts based on all 10 of the analogs. A similar approach was applied for the seven-day WAIP, and with the longer intensity evolutions to detect bifurcation situations about 25 % of the 2010-2014 season forecasts are objectively designated as bifurcations. As in the five-day version, a substantial improvement in intensity mean absolute errors is achieved if an always perfect selection of the correct cluster WAIP forecast of each bifurcation situation is made. Guidance-on-guidance to assist the forecaster by objectively predicting the correct cluster WAIP forecast in each bifurcation situation is being tested. For example, if the Rapid Intensification (RI) index exceeds a critical value, and the analog intensity evolutions also have a reliable indication of RI events, then the track cluster with the larger intensification rate will be selected. Conversely, if the RI index is small and a reliable indication of an RI event is missing in the analog intensity evolutions, the track cluster with the smaller intensification rate will be selected. Predictors in the Statistical Typhoon Intensity Predication Scheme (STIPS) such as vertical wind shear will be utilized to distinquish between non-intensifiers versus intensifiers. The capability of this objective prediction of the correct selection of the intensity cluster will evaluated relative to the always perfect selection. Nevertheless, it will be the forecaster who has the final decision as to which intensity cluster of the bifurcation event should be selected to provide the most effective warning.