Real-Time Identification of Equatorial Waves and Evaluation of Waves in Global Forecasts

Abstract

A novel technique is developed to identify equatorial waves in analyses and forecasts. In a real-time operational context, it is not possible to apply a frequency filter based on a wide centered time window due to the lack of future data. Therefore, equatorial wave identification is performed based primarily on spatial projection onto wave mode horizontal structures. Spatial projection alone cannot distinguish eastward- from westward-moving waves, so a broadband frequency filter is also applied. The novelty in the real-time technique is to off-center the time window needed for frequency filtering, using forecasts to extend the window beyond the current analysis. The quality of this equatorial wave diagnosis is evaluated. First, the “edge effect” arising because the analysis is near the end of the filter time window is assessed. Second, the impact of using forecasts to extend the window beyond the current date is quantified. Both impacts are shown to be small referenced to wave diagnosis based on a centered time window of reanalysis data. The technique is used to evaluate the skill of the Met Office forecast system in 2015–18. Global forecasts exhibit substantial skill (correlation > 0.6) in equatorial waves, to at least day 4 for Kelvin waves and day 6 for westward mixed Rossby–gravity (WMRG), and meridional mode number n = 1 and n = 2 Rossby waves. A local wave phase diagram is introduced that is useful to visualize and validate wave forecasts. It shows that in the model Kelvin waves systematically propagate too fast, and there is a 25% underestimate of amplitude in Kelvin and WMRG waves over the central Pacific.

Comparison of forecasting accuracy for Madden Julian Oscillation (MJO) and Convectively Coupled Equatorial Waves (CCEW) using Tropical Rainfall Measuring Mission (TRMM) and ERA-Interim precipitation forecast data for Indonesia

“…The most recent study by Yang et al (2021) exhibited a higher correlation than the result of this research, that is >0.6 at least day +4 for Kelvin wave and day +6 for MRG and meridional mode number n=1 and n =2 ER wave. Meanwhile, the correlation obtained in this study shows a similar correlation value only at least day +3 for ER and MRG wave by the filtered 3B42-AR(2) process and for all forecast time only for ER by the filtered ERA-Interim precipitation forecast process.…”

“…The references data perhaps become one of the causes of this significant difference. Yang et al (2021) utilized the analysis dataset from the same NWP model as a reference, while this study used a TRMM product as a reference which also being a proxy of surface precipitation. Another reason is due to a different reference parameter.…”

“…Another reason is due to a different reference parameter. Yang et al (2021) examined wind at 850 hPa, while this study examined a rainfall estimation that includes a complex mechanism compare to other atmospheric parameters. However, both studies suggest that the predictability of the Kelvin wave exhibited a weaker forecast skill compare to other wave types.…”

Dynamical propagation and growth mechanisms for convectively coupled equatorial Kelvin waves over the Indian Ocean

“…CCKWs have been analysed in numerical weather predictions (NWPs) and are generally too weak (Dias et al ., 2018; Janiga et al ., 2018). Phase speeds can be either too fast (Dias et al ., 2018; Yang et al ., 2021) or too slow (Janiga et al ., 2018). Again, a dynamical vorticity budget analysis could be applied to NWP output to help further diagnose these systematic errors.…”

“…CCKWs are responsible for high‐impact weather (Ferrett et al ., 2020; Lubis and Respati, 2021), including major flooding incidents (Baranowski et al ., 2020; Latos et al ., 2021). They can be identified and skilfully forecast in real time in atmospheric models (Yang et al ., 2021), but also have a significant, possibly two‐way, interaction with the underlying ocean (Baranowski et al ., 2016a; Baranowski et al ., 2017).…”