1. Introduction
The South China Sea (SCS) and Maritime Continent (MC) regions are located at the center of the Indo-Pacific warm pool. This region is at the heart of the rising branch of both the Hadley circulation and Walker circulation and is also the critical pathway of the Asian–Australian monsoon system (Chang et al. 2005). Convection over MC provides the heat source for driving the extratropical circulation (Ramage 1968; Neale and Slingo 2003), and through teleconnections it can influence weather and climate over remote places including East Asia (Chang and Lau 1982; Lau and Chang 1987), North America (Yanai and Tomita 1998; Yang et al. 2002; Chan and Li 2004), and Europe (Neale and Slingo 2003).
Convection over the SCS–MC region exhibits significant multiscale variability, which remains a great challenge to the global atmospheric models, owing to the difficulties in representing convective processes in the tropical environment with contrasting land–ocean difference, complex coastlines, and steep topography (Birch et al. 2015). This region is surrounded by islands and continents with complex topography, which cultivates prominent diurnal variability of convection. Periodically and zonally propagating modes of tropical convection at different temporal and spatial scales can be found active over the SCS–MC. These are regarded as convectively coupled tropical waves based on the theoretical study of Matsuno (1966) and the analysis of Wheeler and Kiladis (1999), including the Madden–Julian oscillation (MJO; Madden and Julian 1971, 1972, 1994), Kelvin waves (KWs), equatorial Rossby (ER) waves, mixed Rossby–gravity (MRG) waves, and tropical depression (TD)-type disturbances (Takayabu and Nitta 1993).
The convectively coupled equatorial KWs are the leading eastward propagation modes near the equator, with phase speeds around 10–15 m s−1 (Kiladis et al. 2009) and higher amplitudes over the MC (Roundy 2008). The KWs can be embedded in the convective envelope of MJO events as “building blocks” (Nakazawa 1988; Majda et al. 2004; Mapes et al. 2006; Gottschalck et al. 2013) or become active as an independent mode (Dunkerton and Crum 1995; Wheeler and Kiladis 1999). Significant ocean–atmosphere interactions can occur during the passage of the KWs (Baranowski et al. 2016a). The KWs can significantly modulate the tropical convection on synoptic scales (e.g., Takayabu 1991; Wheeler and Kiladis 1999; Wheeler et al. 2000; Wang and Fu 2007; Ventrice et al. 2012) and monsoon onset (e.g., Flatau et al. 2003; Straub et al. 2006). Previous studies also discussed the relationships between KWs and TC genesis over the western Pacific. Schreck and Molinari (2011) examined cases of tropical cyclone (TC) genesis in cyclonic potential vorticity areas created between the superimposed westerlies brought by MJO and KW events and easterly trade winds over the western North Pacific. With the passage of strong KWs, the enhanced westerlies broke potential vorticity areas into vortices, leading to TC genesis. Schreck (2015) found that TC genesis over the western North Pacific (WNP) is most favored in the half days after the passage of KW, based on the statistics of 380 TCs. The composite resembles the result of Schreck and Molinari (2011) that westerlies associated with KWs offer cyclonic vorticity for pre-TC vortex, and the convective phase of the MJO provides a favorable background condition for the interaction of KWs and TC genesis to take place.
Tropical waves can also interact with the prominent diurnal variability over the MC and SCS. In MJO events, the mean and diurnal amplitude of land precipitations over the MC are enhanced 6 days ahead of the MJO convection envelope, while the precipitation over the coastal ocean is largely suppressed (Peatman et al. 2014; Birch et al. 2016; Hung and Sui 2018). Baranowski et al. (2016b) tracked the KW events passing MC using satellite observations to identify the scale interactions between the KWs and the diurnal cycle (DC) over the major islands. For those KWs that arrive in phase with the local DC, the KW-associated precipitation is enhanced by 3 times and the chance of successful KWs traversing the MC is 40% higher, when compared to the KWs that arrive at other times of the day. While the results of Baranowski et al. (2016b) emphasize more the effects of the local DC over the islands on the passing KW, the modulation of the local DC by the KW, which is less explored in previous studies, is also worth investigating.
Large-scale conditions over the MC are also an important modulator of the onset of the Australian summer monsoon (ASM), which climatologically occurs around mid- to late December. The influence of MJO on the onset of ASM has been identified in earlier studies (Hendon and Liebmann 1990; Wheeler and Hendon 2004; Evans et al. 2014). Wheeler and Hendon (2004) investigated the ASM onset over the 28 years from 1974/75 to 2001/02. Over 60% (17 years) of the onset events occurred when the MJO was active [i.e., with the Real-time Multivariate MJO (RMM) index exhibiting amplitude >1], mostly during RMM phases 4–7. When the MJO convective center and associated low-level westerlies appear in the vicinity of northern Australia, they provide favorable conditions for the transition of circulation and convective instability. However, within the broad time range of phases 4–7, or when the MJO is inactive, the actual onset date of the ASM is likely modulated by other phenomena at shorter time scales (Wheeler and McBride 2005). While earlier studies have identified the positive contribution by the KWs to the summer monsoon onset over India (Flatau et al. 2003) and SCS (Straub et al. 2006), the potential influence of the KWs on the ASM onset is less discussed.
Most studies of the synoptic and subseasonal variability in the tropics focused on the activity of the MJO (Waliser et al. 2006; Zhu et al. 2014). The month of December 2016 presents a scenario to study the effect of equatorial convectively coupled KWs over the SCS-MC region, when they were active while the MJO signal was absent. The goal of the present study is to investigate the scale interactions of these KWs, and their impacts on the rainfall variability of other synoptic events, the DC, and the onset of the Australian monsoon. Section 2 describes the data and the methodology. Section 3 presents the background conditions in December 2016. Section 4 describes the identification of the equatorial KWs and other tropical modes in the region. Section 5 discusses the scale interactions involving the KWs during this month, including the interactions with other tropical disturbances, the modulation of DC over MC, and the potential relationship with the onset of the Australian summer monsoon. The summary and concluding remarks are given in section 6.
2. Data and methodology
The datasets used for this study can be categorized into climatology and near-real-time observations. Climatology data are used to produce anomalies on interannual time scales and to filter out tropical waves. Near-real-time data are processed for monitoring weather events.
a. Observation and reanalysis datasets
For the climatological outgoing longwave radiation (OLR) field, we use interpolated data obtained from National Oceanic and Atmospheric Administration (NOAA) polar-orbiting satellite (Liebmann and Smith 1996). It has daily 2.5° × 2.5° resolution and the period between January 1979 and December 2013 was chosen to compute OLR climatology. The OLR for December 2016 is obtained from NOAA Climate Data Record (CDR) of OLR version 1.2 (Lee and NOAA CDR Program 2011), which is estimated from High-Resolution Infrared Radiation Sounder (HIRS) radiance observations with a 2-day lag. It is given daily with 1° × 1° horizontal resolution.
The European Centre for Medium-Range Weather Forecasts interim reanalysis (ERA-Interim, hereinafter ERA-Int; Dee et al. 2011) is utilized for zonal and meridional wind field. The data assimilation system producing ERA-Int is based on integrated forecast system (IFS-Cy31r2, 2006 release). January 1979 to December 2013 was also chosen as the period to compute climatology to match that of the OLR. The data of December 2016 are used to diagnose the synoptic evolution of wind fields and the circulation associated with the filtered waves during this month. The horizontal grid resolution is 0.75° × 0.75° with 60 vertical levels from the surface up to 0.1 hPa. Analysis data are available every 6 h (0000, 0600, 1200, 1800 UTC).
Sea surface temperature (SST) was based on NOAA 0.25° daily Optimum Interpolation Sea Surface Temperature (OISST v2; Banzon et al. 2016), which is an analysis including observations from other platforms such as satellites, ships, and buoys. Here the OISST dataset derived from only the Advanced Very High Resolution Radiometer (AVHRR) is used. The climatology SST is based on the period from December 1981 to December 2015.
We apply the Climate Prediction Center (CPC) morphing technique (CMORPH) satellite precipitation estimates (version 1.0 CRT; Joyce et al. 2004; Xie et al. 2017) as rainfall data. It is derived by combining satellite infrared and microwave sounders, with calibration against surface gauge observations. The temporal and spatial resolution of CMORPH data used in the present study is 3 hourly and 0.25° × 0.25° in the latitude–longitude grid. The period for computing rainfall climatology is January 1998 to December 2015.
b. Interannual anomaly and space–time wave filtering
We calculated the interannual time scale anomaly for December 2016 by removing the long-term climatological mean and the first three harmonics of the annual cycle of the target variables.
To extract the convectively coupled waves from the data of December 2016, we performed wavenumber–frequency filtering (Wheeler and Kiladis 1999) to daily OLR and 925-hPa wind anomaly of this month. The long-term mean and annual cycle were removed to obtain the anomaly as in the first and second step of calculating the interannual anomaly. To retain the full signal of the waves, as Kiladis et al. (2009) did, we do not decompose the anomaly into the symmetric and asymmetric component. The anomaly data in space and time are then Fourier-transformed into zonal wavenumber and frequency domain. Then, the Fourier coefficients falling outside the wavenumber and frequency bands corresponding to each of the tropical modes are set to zero. Tropical modes considered in this study are MJO, KW, ER, and MRG/TD (referred to jointly as MT) waves. The wavenumber and frequency bands for each tropical modes are presented in Table 1. MJO and TD-type disturbance do not fit the linear wave solutions but are of prominent modes in the real world (Madden and Julian 1994; Takayabu and Nitta 1993; Wheeler and Kiladis 1999). The MRG and TD-type can be considered together as the MT band. Takayabu and Nitta (1993) found the transition of MRG to TD-type disturbance near the western Pacific. Frank and Roundy (2006) have found that the TD-type disturbance is the most asymmetric modes among other tropical waves and its amplitude is largely confined in Northern Hemisphere. Furthermore, they also mentioned due to the evolution of MRG to TD-type disturbance, any filter that designed to separate these two waves suffers from contamination of the other wave type (Roundy and Frank 2004; Frank and Roundy 2006). For this reason, we chose MT instead of pure MRG waves (Matsuno 1966) in this study.
The range of planetary zonal wavenumber, period, and equivalent depth chosen for filtering waves and their corresponding reference. Positive (negative) planetary zonal wavenumber indicates eastward (westward) propagation. MJO and MT do not follow the dispersion curve so the equivalent depths are not calculated.
c. Diurnal cycle analysis
To analyze the diurnal variation of precipitation, the dates are separated into convectively active and suppressed conditions over the western part of MC (see section 5c for details). Ling et al. (2019) pointed out that the harmonic analysis may not accurately represent the precipitation DC over the MC, especially over land areas, because the developing phase can be shorter (faster) than the decaying phase in the daily cycle. As the data here only cover a month, the DC phase and amplitude are determined using a simple method. The composite means of the 3-hourly CMORPH precipitation (r) for the two conditions are then calculated at the eight local times at each grid. The diurnal amplitude is the rainfall difference between the composite mean daily maximum and minimum, while the diurnal peak time is the local hour when the composite mean daily maximum occurs.
3. Background conditions in December 2016
The long-term climatological mean OLR and 925-hPa streamline for December (Fig. 1a) are characterized by the convergent circulation into the tropical belt, and convection is confined over the Indian Ocean (IO), MC, western Pacific (WP), intertropical convergence zone (ITCZ), and South Pacific convergence zone (SPCZ). In the Northern Hemisphere (NH), radiative cooling over the Asian continent drives the cold-core Siberian high, bringing the low-level northeasterly winter monsoon along the coast of the Asian continent toward the IO and MC. The convective activity over the MC is the most vigorous in boreal winter because of the monsoon–terrain interactions (Chang et al. 2005). The low-level flow over the IO resembles a Gill response. The cyclonic flow over northwestern Borneo is the Borneo vortex, a prominent synoptic feature in boreal winter (Johnson and Houze 1987). Winds over the WP around 5°–15°N are dominant by northeasterly trade winds. The mean rainfall distribution (Fig. 1b) is consistent with the mean OLR with highest rainfall over coastal areas of SCS-MC, especially for the windward side such as the eastern Philippines, Vietnam, the Malay Peninsula, and western Sumatra, generally as a result of the wind–terrain interactions (Chang et al. 2016). For the SST climatology in December (Fig. 1c), high SST (over 28.5°C) is confined around the equator within 10° latitudes in the IO, MC, and WP, and maximizes in northern Australia.
December climatological mean (a) OLR (W m−2; shaded) and 925-hPa streamlines, (b) CMORPH rainfall (mm day−1), and (c) SST (°C; shaded) and 925-hPa wind vectors (m s−1).
Citation: Journal of Climate 32, 20; 10.1175/JCLI-D-18-0471.1
Figure 2 shows the interannual anomaly of convection, SST, and low-level circulation for December 2016. The ENSO condition for this month is in the La Niña phase following the strong El Niño event in 2015/16. The SST and convection anomaly over the equatorial region therefore exhibited features corresponding to the La Niña condition, including warmer SST by 0.5°–2°C and generally more enhanced convection over the tropical WP, SCS, and MC, as well as colder SST by 0.5°–2°C and suppressed convection over the equatorial IO (60°–100°E).The anomalous Gill-type response in low-level winds over the southeastern IO and SCS is particularly strong this month. Anomalous northwesterlies over the area south of the Sumatra and Java islands (5°–15°S, 95°–110°E) enhanced the climatological northwesterly, which likely led to increased SST by 0.5°–2°C through the coastal Ekman effect. Farther south in the IO (15°–20°S, 95°–110°E) the anomalous southeasterly was also in phase with the climatological southeasterly. The cold SST anomaly over this region is likely owing to the wind evaporation effects. Over the SCS and WNP, the Gill-type response led to interannually anomalous southwesterly winds at 925 hPa, opposing the climatological northeasterly of the winter monsoon.
Interannual component of the monthly anomalous (a) OLR (W m−2), (b) precipitation (mm day−1), and (c) SST (°C; shaded) and 925-hPa wind vectors (m s−1) in December 2016.
Citation: Journal of Climate 32, 20; 10.1175/JCLI-D-18-0471.1
Cold air outbreaks of the East Asia winter monsoon were also not active in December 2016. We examined the cold surge index in Lim et al. (2017), in which the mean wind speed of the 850-hPa northeasterly wind over the SCS (5°–10°N, 107°–115°E) needs to exceed 9.65 m s−1 (0.75 standard deviations above the long-term mean) and the maximum mean sea level pressure (MSLP) over the northern SCS (18°–22°N, 105°–122°E) needs to be above 1020 hPa. Such criteria were never satisfied during December 2016 (figure not shown). Two periods (5–7 and 16–18 December) briefly satisfied the MSLP requirement and exhibited northeasterly conditions, but the areal mean wind speed was only around 6 m s−1, indicating moderate events of a cold air outbreak. Therefore the tropical–extratropical interaction associated with the East Asian cold surge was not significant during this month.
4. Convectively coupled tropical waves in December 2016
a. Kelvin waves
Figure 3a shows the Hovmöller diagram of total OLR and 925-hPa zonal wind from the IO to the WP in the equatorial band (5°S–5°N) during December 2016. There are three distinct large-scale deep convective systems that emerged over the eastern IO, propagated eastward through the MC to the WP, and dissipated at around 160°E. Based on the wavenumber–frequency filtered OLR (Fig. 3b), the three events are identified as convectively coupled KWs (labeled as KW1 to KW3). During KW1, which was short-lived and weak, the low-level westerly wind was less coupled with deep convection. Leading the strong negative OLR anomaly of KW2 and KW3, the low-level westerly wind appeared (3–5 m s−1) and then amplified (7–9 m s−1) while coupled with the convection. KW2 is the strongest case in this month, which was amplified over eastern IO on 10 December, propagated across the southern MC to the WP, and dissipated over 160°E on 20 December. Along the path of KW2, part of the unfiltered OLR is lower than 190 W m−2 and the filtered OLR reached −30 W m−2. KW3 was enhanced over the eastern IO on 26 December and propagated eastward to the tropical WP.
Hovmöller diagram in 5°S–5°N latitude band of (a) total OLR (shaded) superimposed by contours of 925-hPa westerly wind (starts at 3 m s−1; intervals at 2 m s−1; only positive values are shown) and (b) total OLR (shaded) superimposed by wavenumber–frequency filtered OLR for KW (white contour; only negative values are plotted; intervals at −10 W m−2) in December 2016. Black arrows show the propagation of the three KW events based on the linear regression of the filtered OLR contour. The black horizontal bar on the top of (a) is the longitude range for examining diurnal cycle, while the purple (red) vertical lines on the right y axis of (a) mark the date range of convectively active (suppressed) phase over WMC (see Table 3 for detail).
Citation: Journal of Climate 32, 20; 10.1175/JCLI-D-18-0471.1
We can also track the signals of KW2 and KW3 from the sequence of weather maps across December 2016. Figure 4 shows the horizontal evolution of unfiltered OLR and streamline at 925 hPa every two days from 6 to 28 December. KW2 strengthened over the eastern IO on 10 December (Fig. 4c) after the enhanced westerly wind over eastern IO, passing through the southern MC during 12–18 December (Figs. 4b–g) and decayed at 160°E on 20 December. On 26 December, KW3 strengthened over the eastern IO (Fig. 4k) following the enhanced westerly wind over the eastern IO as KW2 did and propagated eastward.
Sequential weather maps of OLR (shaded) and 925-hPa streamline at every two days between 6 and 28 Dec 2016. The white ellipses are filtered OLR anomaly of the KWs. Abbreviations: “V”: TC Vardah; “BV”: Borneo vortex; “TD”: tropical depression; “06U”: TD 06U; “Y”: TC Yvette; “N”: TC Nock-ten.
Citation: Journal of Climate 32, 20; 10.1175/JCLI-D-18-0471.1
The observed wave characteristics (zonal wavelength, period, intrinsic and zonal phase speed) are listed in Table 2. They are in general consistent with previous research (Wheeler and Kiladis 1999; Kiladis et al. 2009). The zonal phase speeds of the observed KWs in this month (11.97–15.44 m s−1) are somewhat slower than the intrinsic one (15.81 m s−1). The spatial map of the filtered OLR and 925-hPa wind of KW2 on 8 December is shown in Fig. 5a. The convective signal is confined near the equator, and the circulation signal is mainly in the zonal wind component. The pattern is consistent with KW linear solution (Matsuno 1966) and previous Kelvin wave composite study (Wheeler et al. 2000; Kiladis et al. 2009).
Observed zonal wavelength, period, and zonal phase speed (m s−1) for strong KW, ER, and MT waves in December 2016. Each case is labeled in Figs. 3 and 6. Intrinsic zonal phase speed is calculated based on 25-m equivalent depth of their theoretical dispersion relation (without basic state; Matsuno 1966).
Filtered OLR (shaded) and 925-hPa wind field (vector) for (a) KW2 (ellipse), (b) ER1 wave train (rectangle) and westerly wind (ellipse), (c) ER2 wave train (rectangle), (d) MT1 and its cyclonic circulation (ellipse), (e) MT2 and MT3 wave train (rectangle), and (f) MT4 (rectangle).
Citation: Journal of Climate 32, 20; 10.1175/JCLI-D-18-0471.1
We note that the MJO mode is insignificant in our filtered results. This is further confirmed by examining the RMM index (Wheeler and Hendon 2004), of which the amplitude of the MJO is mostly lower than 1.0 through the month except marginally over 1.0 on 20 and 21 December (figure not shown). Therefore the KWs are the dominant modes over the equatorial regions between IO and WP in December 2016.
b. ER and MT waves
The spatial distribution of the filtered OLR and low-level wind for the ER and MT band (Figs. 5b–f) shows that their associated variabilities are most significant over the off-equatorial regions, especially in the Northern Hemisphere between 5° and 15°N. These waves can be summarized by the Hovmöller diagram between 5° and 15°N in Fig. 6. The total OLR and 925-hPa zonal wind are dominated by westward-propagating signals. After the wave filtering (contours in Fig. 6), we can identify multiple events of ER and MT waves and then track them on the spatial maps in Fig. 4, sequentially, as follows:
ER1 was initiated around 100°–110°E on 1 December. Based on Figs. 4a–d, as ER1 propagated into the Bay of Bengal (BoB) on 6 December, TC Vardah formed within its envelope (also see Fig. 6) and eventually dissipated over India on 12 December.
MT1 initiated on 6 December around 130°E. It propagated from WP to SCS on 8 December and a Borneo vortex was formed (Fig. 4b) when MT1 terminated over southern SCS (Fig. 6).
MT2 was a weak event, which appeared over the BoB on 16 December and disappeared over the central IO around 20 to 22 December (Figs. 4f–h).
MT3 was initiated strongly on 16 December over eastern Philippine (Fig. 4f). Although filtered OLR of MT3 weakened over 100°E on 20 December (Fig. 6), the associated convection and circulation can be tracked on the weather map continuously as it moved into the BoB (Figs. 4g–j).
ER2 appeared over 150°E on 18 December and its westward-propagating OLR signal weakened after intersecting with that of MT4 around 130°E on 22 December.
MT4 started over 150°E on 20 December and TC Nock-ten formed in its envelope. MT4 and TC Nock-ten eventually decayed over SCS on 28 December (Figs. 4h,i).
As in Fig. 3b, but for the 5°–15°N latitude band. The contours are filtered OLR for ER (red) and MT (dark blue), with solid (dashed) lines indicating negative (positive) values at intervals of 10 W m−2 for ER and 15 W m−2 for MT.
Citation: Journal of Climate 32, 20; 10.1175/JCLI-D-18-0471.1
Table 2 shows that the observed ER and MT waves are faster than the intrinsic mode. The filtered OLR and 925-hPa wind for ER1 (Fig. 5b) and ER2 (Fig. 5c) resemble the n = 2 ER (Matsuno 1966) in linear solution, which has an antisymmetric pattern in its circulation. MT1 (Fig. 5d) and MT4 (Fig. 5f) exhibit a TD-type disturbance (Takayabu and Nitta 1993; Roundy and Frank 2004), and the associated OLR anomaly is mainly confined in NH, with cyclonic (anticyclonic) circulation in phase with negative (positive) OLR, and with less cross-equatorial meridional flow (Fig. 5f). MT2 and MT3 (Fig. 5e) were associated with antisymmetric circulation structures and cross-equatorial meridional flows and their variabilities over northern equator were stronger.
5. Scale interactions involving the KWs
a. The wave–wave interactions
The interactions between the KWs and the off-equatorial waves were identified by combining the equatorial 925-hPa westerly and filtered KW in Fig. 3a with the off-equatorial OLR in the NH in Fig. 6, as shown in Fig. 7. When the KW convective envelope intersects with the off-equatorial wave signals, it indicates that the KW and the westward-propagating waves arrived at the same longitude. Although the OLR variabilities of the westward-propagating waves were mostly confined in the off-equatorial region, the anomalous low-level winds can still influence the equatorial region, as shown in Fig. 5, and their contribution to the equatorial westerly can be identified in the Hovmöller diagram of Fig. 7. To confirm the occurrence of interaction and to obtain the detailed evolution, the location and propagation of the wave packets, enhanced wind pattern, and synoptic disturbances around the intersecting period were traced in the weather maps of Fig. 4. Following this procedure, four events of scale interactions involving the KWs were identified in December 2016.
A schematic figure explaining extratropical–tropical interactions in December 2016. The thin black contours are u925 in 5°S–5°N (identical to Fig. 6), and the shading is the total OLR in 5°–15°N (identical to shading in Fig. 6). The smooth ellipses represent wave events based on the filtered OLR in Figs. 3b and 6 (blue: KW, red: ER, dark blue: MT). The numbered star symbols indicate the events of wave–wave interactions: ☆I1: TC Vardah in ER1 enhanced southwesterly winds and coastal convection over western Sumatra → KW2 enhanced. ☆I2: MT1 over SCS → Borneo vortex + KW2 equatorial westerly → TD. ☆I3: KW2 + WP trade wind + ER2 → MT4/TC Nock-ten. ☆I4: MT3 → enhanced westerlies and convection over western Sumatra → terrain effect of Sumatra → KW3 enhanced.
Citation: Journal of Climate 32, 20; 10.1175/JCLI-D-18-0471.1
The first interaction (I1) occurred around 8–10 December over the eastern IO (80°–90°E) where equatorial westerlies/southwesterlies were enhanced by TC Vardah embedded in ER1 (see Figs. 4a,b and 5b). The enhanced equatorial westerlies over the terrain of Sumatra favored the development of organized convection over the west coast of Sumatra on 8 December. As the convective center of KW2 approached Sumatra on 10 December, the convective signal of KW2 was strengthened afterward (Figs. 3b and 4c–e). This enhancement likely helped KW2 maintain a longer lifetime and propagated farther east.
The second interaction (I2) occurred also during 8–10 December but over the SCS. As mentioned in section 3, northeasterly from moderate cold air outbreak entered SCS on 5–7 December (Fig. 4a). With the terrain effect of the southern SCS basin, a Borneo vortex was formed on 8 December (Fig. 4b) when MT1 contributed to enhanced northeasterly winds over west coast of Borneo (Fig. 5d). As the equatorial westerlies accompanied with KW2 propagated into MC on 10 December, the Borneo vortex further intensified into a TD owing to the enhanced shear vorticity (Figs. 4c,d). The northeasterly wind from the cold air outbreak was not maintained during the intensification of the Borneo vortex, indicating that the extratropical–tropical interaction did not play a significant role with regard to I2. The unnamed TD eventually made landfall at Vietnam on 14 December (Fig. 4e).
The third interaction (I3) occurred during 18–20 December. As KW2 propagated to the WP (150°E) on 18 December, the associated equatorial westerlies and the easterly trade wind produced low-level cyclonic circulation, with enhanced convection (L in Fig. 4g). The positive vorticity area over WP sustained, with the contribution also from the circulation of ER2, and developed into MT4 and TC Nock-ten on 20–22 December (Figs. 4h,i and 5f), which propagated westward and made landfall at the Philippines on 26 December (Fig. 4k), and then weakened after entering the SCS. In the case study in Schreck and Molinari (2011), two TCs developed in the breakdown of cyclonic potential vorticity formed in the strip between equatorial westerly and easterly trade winds. In their cases, the westerlies were enhanced by the MJO and KWs together, and the MJO also provided the convective envelope. In the case of I3 here the westerlies and convection were mainly contributed by the KWs.
The last interaction (I4) occurred during 20–26 December. The cyclonic circulation of MT3 led to the equatorial westerlies onshore of the west coast of Sumatra on 20–24 December (Figs. 4h–j), accompanied with sustained coastal convection that was then coupled to the circulation of KW3 and lead to the intensification of KW3 after 26 December (Figs. 4k and 3a). The evolution resembled the intensification of KW2 in I1, except that the westerly winds and convection in the west of Sumatra was induced by MT instead of ER.
b. Diurnal cycle in active and suppressed phases of KWs
To analyze the modulation of the diurnal cycle of MC by the KWs in December 2016, the dates were chosen and composited according to the daily areal mean OLR over the western part of the MC (WMC; 7.5°S–7.5°N, 95°–120°E). The days when the daily areal mean OLR is 0.5 standard deviations below the monthly mean (<187.7 W m−2) represented the condition that the WMC was convectively active and the eastern MC (EMC; 7.5°S–7.5°N, 120°–150°E) was convectively suppressed. Conversely, the days with mean OLR above the monthly mean by 0.5 standard deviations (>200.2 W m−2) represented the condition that the WMC was suppressed and the EMC was convectively active. The dates composited in each condition are listed in the footnote of Table 3 and marked on the vertical axis of Fig. 3a on the right-hand side, and they are primarily dominated by the phases of KWs. Figures 8a and 8d show the composite mean total precipitation and low-level winds of the two scenarios, respectively. The convectively active region exhibited extensive coverage of precipitation and strong equatorial westerlies, while the suppressed region exhibited limited and scattered precipitation and weak wind condition.
Composite mean precipitation and diurnal cycle amplitudes over WMC and EMCa for the convectively active and suppressed conditions of the KWs.b
Composite mean (a) CMORPH precipitation (shaded; mm day−1) and 925-hPa wind vectors (m s−1), (b) diurnal precipitation amplitude (shaded; mm day−1), and (c) diurnal peak time (local hour) over areas with diurnal amplitude ≥24 mm day−1 when WMC exhibits convectively active conditions and EMC exhibits suppressed conditions. (d)–(f) As in (a)–(c), but showing the composite mean when WMC exhibits suppressed conditions and EMC exhibits active conditions. Vectors in (b) and (e) are 925-hPa anomalous wind relative to the monthly mean of December 2016. The convectively active and suppressed phases are defined in Table 3. The left and right rectangle boxes in (a) and (d) as well as the dashed vertical dashed lines in (b) and (e) mark the areas of WMC and EMC.
Citation: Journal of Climate 32, 20; 10.1175/JCLI-D-18-0471.1
Figure 8b presents the amplitude of the composite DC of precipitation for the convectively active WMC and suppressed EMC, as well as the composite anomaly of 925-hPa wind. There was an anomalous cyclonic circulation over the southern SCS/western Borneo, and DC was active over most of the coastal ocean. DC amplitude is strong especially the northwest coast of Borneo, as well as on the islands over southwestern Borneo and southern Sumatra. Over the suppressed EMC there was a strong anomalous anticyclonic circulation of WP/northwestern New Guinea. DC was only active over the New Guinea Island and the nearshore coastal ocean, as well as a limited area over WP around 140°E. The distribution of the DC amplitude generally reflected the composite total value in both active and suppressed conditions. Figure 8c shows the DC peak time of the same composite over locations with DC amplitude ≥ 24 mm day−1. Over land, the peak time in mountainous areas occurred mostly in the late afternoon to early evening (1400–2000 LT), and at midnight (0000–0400 LT) in some downwind plain areas. Over the coastal ocean to the west of Borneo, a progressive phase shift in DC can be found, with peak times at midnight (0000–0400 LT) near shore and in the early morning (0400–1000) offshore, features the “seaside coastal regime” described by Kikuchi and Wang (2008).
Figures 8e and 8f present the DC and anomalous low-level winds for the condition of convectively suppressed WMC and active EMC. The ocean in the north of the New Guinea island was under the anomalous cyclonic flow, and strong DC occurred extensively from coastal areas to open ocean, also with the progressive peak time from midnight to early morning. The DC over the New Guinea island was also active. Over the Banda Sea (0°–5°S) there were strong anomalous westerly winds and enhanced composite mean precipitation, but the contribution from the DC was limited. Over the WMC, a weak anomalous anticyclonic circulation and diminished precipitation were seen over southern SCS. There were strong anomalous northeasterly winds over eastern Borneo Island and the Java Sea, the only regions with active DC in the suppressed WMC.
Table 3 listed the mean DC amplitude over land and ocean in WMC and EMC for the two composites. Overall the DC was enhanced over the convectively active region (22.5–24.1 mm day−1), and weakened over the suppressed region (13.5–15.6 mm day−1), without noticeable land–ocean contrast. The response over the ocean is mainly over specific coastal regions (i.e., west of Borneo and north of New Guinea). The enhanced coastal organized convection may play an important role during the scale interactions mentioned in the last section, such as the intensification of the Borneo vortex in I2, and the development of MT4 in I3. Over the land, locations with active DC shifted between suppressed and active condition, with the emergence of specific “hot spots,” likely in response to the terrain effects under different background wind. The pattern of DC under convectively active condition over Borneo island and southern SCS resembled the characteristics of the “strong westerly” regime reported by Ichikawa and Yasunari (2006), including the stronger amplitude over the northwestern part of the island and the coastal propagating signals with enhanced cyclonic circulation associated with the Borneo vortex. Over southern Sumatra, the DC pattern is consistent with the westerly wind regime in Yanase et al. (2017): convection is initiated in the afternoon in the central mountain range and then propagated eastward to the eastern plain in the late evening by the low-level westerly winds. However, we acknowledge here that the sampling numbers in the above DC analysis are limited (9 and 11 days in each composite), so the results should be interpreted more qualitatively and are subject to the interannual conditions of this year. A more extensive climatological analysis on this topic covering KW events in multiple years will be carried out in the future.
c. The KW and Australian summer monsoon onset
The Australian summer monsoon (ASM) onset can be defined using two commonly used indices, and Fig. 9 shows the time series of these indices from mid-November 2016 through the end of January 2017. Hung and Yanai (2004) used the OLR and 850-hPa zonal wind averaged over northern Australia and the Arafura Sea (2°–15°S, 115°–150°E). The onset day is defined as the first day when averaged 850-hPa zonal wind exceeds 2 m s−1 and sustains for longer than 10 days, while the OLR is lower than 210 W m−2 for at least several days during the 10-day period. These criteria were satisfied on 15 December 2016 (red and blue lines in Fig. 9). Kajikawa et al. (2010) used the 850-hPa zonal wind averaged over the southern MC to northern Australia (5°–15°S, 110°–130°E) as the multi-time-scale Australian summer monsoon index (AUSMI). The onset day based on this index features the first day after 1 November when the AUSMI turns positive and sustains for longer than 5 days, provided that the AUSMI must be positive in at least three pentads out of the subsequent four pentads, and the accumulative four-pentad mean AUSMI > 1 m s−1. For 2016/2017 the AUSMI signified a sharp and clear onset on 13 December 2016 (brown line in Fig. 9).
Time series of the two Australian summer monsoon indices from 16 Nov 2016 to 30 Jan 2017. The red and blue lines are OLR and 850-hPa zonal wind, respectively, averaged over northern Australia and the Arafura Sea (2°–15°S, 115°–150°E), used by Hung and Yanai (2004). The brown line is the AUSMI (850-hPa zonal wind averaged over 5°–15°S, 10°–130°E) used by Kajikawa et al. (2010). The green shading marks the onset period (see text in section 5c for definition).
Citation: Journal of Climate 32, 20; 10.1175/JCLI-D-18-0471.1
Both indices consistently indicate the ASM onset in 2016/17 took place around 13–15 December (green shading in Fig. 9). The timing coincided with the passage of KW2 over the MC. Before onset, the large-scale conditions over northern Australia were prevailing easterly winds and convectively suppressed (Figs. 4a–d). As the convective center of KW2 moved from the WMC to the EMC during 14–16 December. (Figs. 4e,f), the accompanied equatorial westerlies were enhanced, with convection developed particularly south of the equator over the coastal ocean of southern MC and then eastward to the Arafura Sea. The interannually warmer SST induced by the Gill-type low-level wind response associated with the La Niña condition (Fig. 2c) over this area also provided favorable conditions for convection. The monsoon trough was established afterward, maintaining the low-level westerly and organized convection over northern Australia, which provided favorable conditions leading to the formation of TD 06U and TC Yvette on 18 December (Figs. 4g–j).
6. Summary and concluding remarks
This study examined the scale interactions over the SCS and MC region involving the tropical convectively coupled Kelvin waves in December 2016, a month during which the KWs are the dominant mode over the equator. From the interannual perspective, this month was in a La Niña condition, with interannually stronger convection over the MC and weaker northeasterly Asian winter monsoon. The space–time filtering of tropical wave modes revealed that the OLR variability over the equatorial bands were mainly contributed by three eastward propagating KW events, while the MJO signal was insignificant through this month. The first event was weak and less coupled. The second KW event (KW2), with the leading strong equatorial westerly coupled to deep convection on 8–10 December over the eastern IO, was the strongest and most long-lasting one among the three. The last event (KW3) of the month was initiated and coupled on 22–24 December also over eastern IO. Both KW2 and KW3 weakened and terminated around 150°–160°E. Prominent westward propagating variabilities over the off-equatorial regions in the Northern Hemisphere (5°–15°N) were identified as the equatorial Rossby waves and mixed Rossby–gravity/TD-type waves.
By superimposing the evolution of the KWs with the westward propagating waves, four events of interactions over the SCS-MC region were identified. In two of these interactions (I1 and I4), the circulation of the off-equatorial waves (ER in I1 and MT in I4) strengthened the equatorial westerly wind onshore of the west coast of Sumatra and led to more active coastal convections. The coupling between the coastal convections and the approaching KWs, which is likely a key process in these interaction events, resulted in the enhancements of the KW2 and KW3. The enhanced KW2 then contributed to the formation of two tropical disturbances. The other two wave–wave interaction events (I2 and I3) led to formation of TD/TC. In I2, the Borneo vortex over SCS was intensified into an unnamed TD as the northeasterly winds associated with the MT1 and the equatorial westerly winds associated with the KW2 enhanced the shear vorticity. In I3, as KW2 propagated to the WP, the convergence of the equatorial westerlies with the northeasterly trade winds, as well as the cyclonic circulation of ER2, provided favorable conditions for the development of MT4 and later TC Nock-ten that made landfall at the Philippines.
As the KWs propagated through the MC, the local diurnal cycle of precipitation over both land and ocean was enhanced in the region within the anomalous cyclonic flow associated with the KW convection center. The ocean response to KW was mostly in the appearance of the “coastal regime” during the convectively active condition, in which the DC peak time varied progressively from midnight at the near shore to morning in the off shore/open ocean. The west coast of the Borneo Island and the north coast of the New Guinea Island were the most responding areas in the WMC and EMC, respectively, and the enhanced coastal convection organization may also provide a favorable condition for the development of the disturbances during the scale interactions of I2 and I3. The modulation of DC over land by the KW convectively active/suppressed conditions was a shift in the DC hot spots depending on the different background winds and the terrain effects, in addition to the enhanced amplitude.
Last, the passage of KW2 over MC coincided and contributed positively to the onset of the Australian summer monsoon, which occurred around 13–15 December 2016. The KW2 provided the low-level westerlies and initial convections over the southern coasts of MC and the Arafura Sea. The large-scale westerlies, the organized convection, and the monsoon trough quickly were established and maintained, likely facilitated also by the interannually warmer SST over this region. The onset was followed by the formation of TD 06U and TC Yvette within the positive vorticity of the monsoon trough areas.
Our analysis for December 2016 showcases that the convectively coupled Kelvin waves can participate in various types of scale interactions in the SCS and MC, and lead to formation and intensification of synoptic events in the tropics and monsoonal regions. With the absence of the MJO signal, the KWs and the associated scale interactions provide useful references for weather monitoring and forecast of this region. We also note that the diurnal coastal convection over specific regions can play a key role in these scale interactions, owing to the shape of the coastlines and orientation of the terrain. The comparable size of the major islands/ocean basins in the SCS-MC area to the scale of the anomalous horizontal circulation of the KWs may also be a factor (Baranowski et al. 2016b). The west coast of Sumatra can be a favorable location of KW enhancement with onshore low-level winds, while over the southern SCS/western coast of Borneo, northern coast of New Guinea, and southern coast of Java coastal organized convection becomes active under synoptic-scale cyclonic circulation and thus potentially favors the intensification of synoptic disturbances. The analysis of the diurnal cycle in the present study only covered a month. In the future, a more extensive climatological analysis focusing on the responses of regional DC to the passage of the KWs can be carried out to obtain more robust features, and the DC cycle can also be identified using the more sophisticated method suggested by Ling et al. (2019). Also, in the present study the observed phase speeds of the tropical waves are computed by the simple linear regression of filtered OLR anomaly. Ling et al. (2014) and Zhang and Ling (2017) provided a more advanced objective tracking method for MJO based on statistics of precipitation anomaly. It is worth exploring how a similar procedure can be developed and applied to track the other wave types in the future.
The influences of scale interactions and tropical–extratropical interactions on regional precipitation over the SCS and MC are the key scientific themes of two international field campaigns that took place after 2016, namely, the Years of the Maritime Continent (YMC, 2017–19; science plan: http://www.jamstec.go.jp/ymc/docs/YMC_SciencePlan_v2.pdf) and the South China Sea Two-Island Monsoon Experiment (SCSTIMX, 2016–19; Lin et al. 2016; https://scstimx.as.ntu.edu.tw/). In the future, analyses similar to those in the present study will be carried out for the winter of 2017 and 2018 to investigate the interannual variability, with the incorporation of the intensive measurements from these two field campaigns.
Acknowledgments
This work is supported by Ministry of Science and Technology of Taiwan (MOST107-2119-M-002-053; WT Chen: MOST107-2119-M-002-024 and MOST108-2111-M-002-004; SP Hsu, CH Sui, and YH Tsai: MOST105-2119-M-002-025, MOST106-2111-M-002-003-MY2, and MOST108-2111-M-002-016). We thank Professor Chi-Pei Chang and the three anonymous reviewers for the insightful comments on the manuscript. The observation and reanalysis data sets were downloaded from the following sources: NOAA Interpolated OLR (1979–2013; https://www.esrl.noaa.gov/psd/data/gridded/data.interp_OLR.html; accessed 15 December 2018), NOAA CDR OLR v1.2 (December 2016; https://www.esrl.noaa.gov/psd/data/gridded/data.olrcdr.interp.html; accessed 15 December 2018), NOAA OISST v2 (1981–2015 and December 2016; https://www.esrl.noaa.gov/psd/data/gridded/data.noaa.oisst.v2.highres.html; accessed 15 December 2018), ERA-Int (1979–2016; https://apps.ecmwf.int/datasets/data/interim-full-daily/levtype=sfc/; accessed 9 May 2019), and CMORPH v1.0 CRT 3-hourly precipitation data (1998–2015 and December 2016;
REFERENCES
Banzon, V., T. M. Smith, T. M. Chin, C. Liu, and W. Hankins, 2016: A long-term record of blended satellite and in situ sea-surface temperature for climate monitoring, modeling and environmental studies. Earth Syst. Sci. Data, 8, 165–176, https://doi.org/10.5194/essd-8-165-2016.
Baranowski, D. B., M. K. Flatau, P. J. Flatau, and A. J. Matthews, 2016a: Impact of atmospheric convectively coupled equatorial Kelvin waves on upper ocean variability. J. Geophys. Res., 121, 2045–2059, https://doi.org/10.1002/2015JD024150.
Baranowski, D. B., M. K. Flatau, P. J. Flatau, and A. J. Matthews, 2016b: Phase locking between atmospheric convectively coupled equatorial Kelvin waves and the diurnal cycle of precipitation over the Maritime Continent. Geophys. Res. Lett., 43, 8269–8276, https://doi.org/10.1002/2016GL069602.
Birch, C. E., M. J. Roberts, L. Garcia-Carreras, D. Ackerley, M. J. Reeder, A. P. Lock, and R. Schiemann, 2015: Sea breeze dynamics and convection initiation: The influence of convective parameterization in weather and climate model biases. J. Climate, 28, 8093–8108, https://doi.org/10.1175/JCLI-D-14-00850.1.
Birch, C. E., S. Webster, S. C. Peatman, D. J. Parker, A. J. Matthews, Y. Li, and M. E. E. Hassim, 2016: Scale interactions between the MJO and the western Maritime Continent. J. Climate, 29, 2471–2492, https://doi.org/10.1175/JCLI-D-15-0557.1.
Chan, J. C. L., and C. Y. Li, 2004: The East Asian winter monsoon. East Asian Monsoon, C.-P. Chang, Ed., World Scientific Series on Asia-Pacific Weather and Climate, Vol. 2, World Scientific, 54–106.
Chang, C.-P., and K. M. Lau, 1982: Short-term planetary-scale interactions over the tropics and midlatitudes during northern winter. Part I: Contrasts between active and inactive periods. Mon. Wea. Rev., 110, 933–946, https://doi.org/10.1175/1520-0493(1982)110<0933:STPSIO>2.0.CO;2.
Chang, C.-P., C.-H. Liu, and H.-C. Kuo, 2003: Typhoon Vamei: An equatorial tropical cyclone formation. Geophys. Res. Lett., 30, 1150, https://doi.org/10.1029/2002GL016365.
Chang, C.-P., Z. Wang, J. McBride, and C. H. Liu, 2005: Annual cycle of Southeast Asia–Maritime Continent rainfall and the asymmetric monsoon transition. J. Climate, 18, 287–301, https://doi.org/10.1175/JCLI-3257.1.
Chang, C.-P., M. Lu, and H. Lim, 2016: Monsoon convection in the Maritime Continent: Interaction of large-scale motion and complex terrain. Meteor. Monogr., Amer. Meteor. Soc., No. 56, 6.1–6.29, https://doi.org/10.1175/AMSMONOGRAPHS-D-15-0011.1.
Dee, D. P., and Coauthors, 2011: The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553–597, https://doi.org/10.1002/qj.828.
Dunkerton, T. J., and F. X. Crum, 1995: Eastward propagating 2- to 15-day equatorial convection and its relation to the tropical intraseasonal oscillation. J. Geophys. Res., 100, 25 781–25 790, https://doi.org/10.1029/95JD02678.
Evans, S. M., R. T. Marchand, and T. P. Ackerman, 2014: Variability of the Australian monsoon and precipitation trends at Darwin. J. Climate, 27, 8487–8500, https://doi.org/10.1175/JCLI-D-13-00422.1.
Flatau, M. K., P. J. Flatau, J. Schmidt, and G. N. Kiladis, 2003: Delayed onset of the 2002 Indian monsoon. Geophys. Res. Lett., 30, 1768, https://doi.org/10.1029/2003GL017434.
Frank, W. M., and P. E. Roundy, 2006: The role of tropical waves in tropical cyclogenesis. Mon. Wea. Rev., 134, 2397–2417, https://doi.org/10.1175/MWR3204.1.
Gottschalck, J., P. E. Roundy, C. J. Schreck, A. Vintzileos, and C. Zhang, 2013: Large-scale atmospheric and oceanic conditions during the 2011–2012 DYNAMO field campaign. Mon. Wea. Rev., 141, 4173–4196, https://doi.org/10.1175/MWR-D-13-00022.1.
Hendon, H. H., and B. Liebmann, 1990: A composite study of the onset of the Australian summer monsoon. J. Atmos. Sci., 47, 2227–2240, https://doi.org/10.1175/1520-0469(1990)047<2227:ACSOOO>2.0.CO;2.
Hung, C.-S., and C.-H. Sui, 2018: A diagnostic study of the evolution of the MJO from Indian Ocean to Maritime Continent: Wave dynamics versus advective moistening processes. J. Climate, 31, 4095–4115, https://doi.org/10.1175/JCLI-D-17-0139.1.
Hung, C.-W., and M. Yanai, 2004: Factors contributing to the onset of the Australian summer monsoon. Quart. J. Roy. Meteor. Soc., 130, 739–758, https://doi.org/10.1256/qj.02.191.
Ichikawa, H., and T. Yasunari, 2006: Time–space characteristics of diurnal rainfall over Borneo and surrounding oceans as observed by TRMM-PR. J. Climate, 19, 1238–1260, https://doi.org/10.1175/JCLI3714.1.
Johnson, R. H., and R. A. Houze Jr., 1987: Precipitating cloud systems of the Asian monsoon. Monsoon Meteorology, C.-P. Chang and T. N. Krishnamurti, Eds., Oxford University Press, 298–353.
Joyce, R. J., J. E. Janowiak, P. A. Arkin, and P. Xie, 2004: CMORPH: A method that produces global precipitation estimates from passive microwave and infrared data at high spatial and temporal resolution. J. Hydrometeor., 5, 487–503, https://doi.org/10.1175/1525-7541(2004)005<0487:CAMTPG>2.0.CO;2.
Kajikawa, Y., B. Wang, and J. Yang, 2010: A multi-time scale Australian monsoon index. Int. J. Climatol., 30, 1114–1120, https://doi.org/10.1002/joc.1955.
Kikuchi, K., and B. Wang, 2008: Diurnal precipitation regimes in the global tropics. J. Climate, 21, 2680–2696, https://doi.org/10.1175/2007JCLI2051.1.
Kiladis, G. N., M. C. Wheeler, P. T. Haertel, K. H. Straub, and P. E. Roundy, 2009: Convectively coupled equatorial waves. Rev. Geophys., 47, RG2003, https://doi.org/10.1029/2008RG000266.
Lau, K. M., and C.-P. Chang, 1987: Planetary scale aspects of winter monsoon and teleconnections. Monsoon Meteorology, C.-P. Chang and T. N. Krishnamurti, Eds., Oxford University Press, 161–202.
Lee, H. T., and NOAA CDR Program, 2011: NOAA Climate Data Record (CDR) of Daily Outgoing Longwave Radiation (OLR), Version 1.2. NOAA National Climatic Data Center, accessed 15 December 2018, https://doi.org/10.7289/V5SJ1HH2.
Liebmann, B., and C. A. Smith, 1996: Description of a complete (interpolated) outgoing longwave radiation dataset. Bull. Amer. Meteor. Soc., 77, 1275–1277.
Lim, S. Y., C. Marzin, P. Xavier, C.-P. Chang, and B. Timbal, 2017: Impacts of boreal winter monsoon cold surges and the interaction with MJO on Southeast Asia rainfall. J. Climate, 30, 4267–4281, https://doi.org/10.1175/JCLI-D-16-0546.1.
Lin, P.-H., and Coauthors, 2016: Interaction of convection over the Maritime Continent–SCS with large-scale flow: 2016 Winter Monsoon Pre-experiment (in Chinese). Atmos. Sci., 44, 237–352.
Ling, J., P. Bauer, P. Bechtold, A. Beljaars, R. Forbes, F. Vitart, M. Ulate, and C. D. Zhang, 2014: Global versus local MJO forecast skill of the ECMWF model during DYNAMO. Mon. Wea. Rev., 142, 2228–2247, https://doi.org/10.1175/MWR-D-13-00292.1.
Ling, J., C. Zhang, R. Joyce, P.-P. Xie, and G. Chen, 2019: Possible role of the diurnal cycle in land convection in the barrier effect on the MJO by the Maritime Continent. Geophys. Res. Lett., 46, 3001–3011, https://doi.org/10.1029/2019GL081962.
Madden, R. A., and P. R. Julian, 1971: Detection of a 40–50 day oscillation in the zonal wind in the tropical Pacific. J. Atmos. Sci., 28, 702–708, https://doi.org/10.1175/1520-0469(1971)028<0702:DOADOI>2.0.CO;2.
Madden, R. A., and P. R. Julian, 1972: Description of global-scale circulation cells in the tropics with a 40–50 day period. J. Atmos. Sci., 29, 1109–1123, https://doi.org/10.1175/1520-0469(1972)029<1109:DOGSCC>2.0.CO;2.
Madden, R. A., and P. R. Julian, 1994: Observations of the 40–50 day tropical oscillation: A review. Mon. Wea. Rev., 122, 814–837, https://doi.org/10.1175/1520-0493(1994)122<0814:OOTDTO>2.0.CO;2.
Majda, A. J., B. Khouider, G. N. Kiladis, K. H. Straub, and M. G. Shefter, 2004: A model for convectively coupled tropical waves: Nonlinearity, rotation, and comparison with observations. J. Atmos. Sci., 61, 2188–2205, https://doi.org/10.1175/1520-0469(2004)061<2188:AMFCCT>2.0.CO;2.
Mapes, B., S. Tulich, J. Lin, and P. Zuidema, 2006: The mesoscale convection life cycle: Building block or prototype for large-scale tropical waves? Dyn. Atmos. Oceans, 42, 3–29, https://doi.org/10.1016/j.dynatmoce.2006.03.003.
Matsuno, T., 1966: Quasi-geostrophic motions in the equatorial area. J. Meteor. Soc. Japan, 44, 25–43, https://doi.org/10.2151/jmsj1965.44.1_25.
Nakazawa, T., 1988: Tropical super clusters within intraseasonal variations over the western Pacific. J. Meteor. Soc. Japan, 66, 823–839, https://doi.org/10.2151/jmsj1965.66.6_823.
Neale, R., and J. Slingo, 2003: The Maritime Continent and its role in the global climate: A GCM study. J. Climate, 16, 834–848, https://doi.org/10.1175/1520-0442(2003)016<0834:TMCAIR>2.0.CO;2.
Peatman, S. C., A. J. Matthews, and D. P. Stevens, 2014: Propagation of the Madden-Julian Oscillation through the Maritime Continent and scale interaction with the diurnal cycle of precipitation. Quart. J. Roy. Meteor. Soc., 140, 814–825, https://doi.org/10.1002/QJ.2161.
Ramage, C. S., 1968: Role of a tropical “maritime continent” in the atmospheric circulation. Mon. Wea. Rev., 96, 365–370, https://doi.org/10.1175/1520-0493(1968)096<0365:ROATMC>2.0.CO;2.
Roundy, P. E., 2008: Analysis of convectively coupled Kelvin waves in the Indian Ocean MJO. J. Atmos. Sci., 65, 1342–1359, https://doi.org/10.1175/2007JAS2345.1.
Roundy, P. E., and W. M. Frank, 2004: A climatology of waves in the equatorial region. J. Atmos. Sci., 61, 2105–2132, https://doi.org/10.1175/1520-0469(2004)061<2105:ACOWIT>2.0.CO;2.
Schreck, C. J., 2015: Kelvin waves and tropical cyclogenesis: A global survey. Mon. Wea. Rev., 143, 3996–4011, https://doi.org/10.1175/MWR-D-15-0111.1.
Schreck, C. J., and J. Molinari, 2011: Tropical cyclogenesis associated with Kelvin waves and the Madden–Julian oscillation. Mon. Wea. Rev., 139, 2723–2734, https://doi.org/10.1175/MWR-D-10-05060.1.
Straub, K. H., G. N. Kiladis, and P. E. Ciesielski, 2006: The role of equatorial waves in the onset of the South China Sea summer monsoon and the demise of El Niño during 1998. Dyn. Atmos. Oceans, 42, 216–238, https://doi.org/10.1016/j.dynatmoce.2006.02.005.
Takayabu, Y. N., 1991: The structure of super cloud clusters observed in 1–20 June 1986 and their relationship to easterly waves. J. Meteor. Soc. Japan, 69, 105–125, https://doi.org/10.2151/jmsj1965.69.1_105.
Takayabu, Y. N., and T. Nitta, 1993: 3-5 day-period disturbances coupled with convection over the tropical Pacific Ocean. J. Meteor. Soc. Japan, 71, 221–246, https://doi.org/10.2151/jmsj1965.71.2_221.
Ventrice, M. J., C. D. Thorncroft, and M. A. Janiga, 2012: Atlantic tropical cyclogenesis: A three-way interaction between an African easterly wave, diurnally varying convection, and a convectively coupled atmospheric Kelvin wave. Mon. Wea. Rev., 140, 1108–1124, https://doi.org/10.1175/MWR-D-11-00122.1.
Waliser, D., and Coauthors, 2006: The Experimental MJO Prediction Project. Bull. Amer. Meteor. Soc., 87, 425–431.
Wang, H., and R. Fu, 2007: The influence of Amazon rainfall on the Atlantic ITCZ through convectively coupled Kelvin waves. J. Climate, 20, 1188–1201, https://doi.org/10.1175/JCLI4061.1.
Wheeler, M. C., and G. N. Kiladis, 1999: Convectively coupled equatorial waves: Analysis of clouds and temperature in the wavenumber–frequency domain. J. Atmos. Sci., 56, 374–399, https://doi.org/10.1175/1520-0469(1999)056<0374:CCEWAO>2.0.CO;2.
Wheeler, M. C., and H. H. Hendon, 2004: An all-season real-time multivariate MJO index: Development of an index for monitoring and prediction. Mon. Wea. Rev., 132, 1917–1932, https://doi.org/10.1175/1520-0493(2004)132<1917:AARMMI>2.0.CO;2.
Wheeler, M. C., and J. L. McBride, 2005: Australian-Indonesian monsoon. Intraseasonal Variability in the Atmosphere–Ocean Climate System, W. K. M. Lau and D. E. Waliser, Eds., Springer, 125–173.
Wheeler, M. C., G. N. Kiladis, and P. J. Webster, 2000: Large-scale dynamical fields associated with convectively coupled Kelvin waves. J. Atmos. Sci., 57, 613–640, https://doi.org/10.1175/1520-0469(2000)057<0613:LSDFAW>2.0.CO;2.
Xie, P., R. Joyce, S. Wu, S.-H. Yoo, Y. Yarosh, F. Sun, and R. Lin, 2017: Reprocessed, bias-corrected CMORPH global high-resolution precipitation estimates from 1998. J. Hydrometeor., 18, 1617–1641, https://doi.org/10.1175/JHM-D-16-0168.1.
Yanai, M., and T. Tomita, 1998: Seasonal and interannual variability of atmospheric heat sources and moisture sinks as determined from NCEP–NCAR reanalysis. J. Climate, 11, 463–482, https://doi.org/10.1175/1520-0442(1998)011<0463:SAIVOA>2.0.CO;2.
Yanase, A., K. Yasunaga, and H. Masunaga, 2017: Relationship between the direction of diurnal rainfall migration and the ambient wind over the southern Sumatra Island. Earth Space Sci., 4, 117–127, https://doi.org/10.1002/2016EA000181.
Yang, S., K. M. Lau, and K. M. Kim, 2002: Variations of the East Asian jet stream and Asian–Pacific–American winter climate anomalies. J. Climate, 15, 306–325, https://doi.org/10.1175/1520-0442(2002)015<0306:VOTEAJ>2.0.CO;2.
Zhang, C. D., and J. Ling, 2017: Barrier effect of the Indo-Pacific Maritime Continent on the MJO: Perspectives from tracking MJO precipitation. J. Climate, 30, 3439–3459, https://doi.org/10.1175/JCLI-D-16-0614.1.
Zhu, H., M. C. Wheeler, A. H. Sobel, and D. Hudson, 2014: Seamless precipitation prediction skill in the tropics and extratropics from a global model. Mon. Wea. Rev., 142, 1556–1569, https://doi.org/10.1175/MWR-D-13-00222.1.
