1. Introduction
In coastal areas, precipitation generation mechanisms differ from those in open-ocean and inland areas because of factors specific to the land–ocean interface, such as surface heating differences and orographic effects of coastal mountain ranges. Coastal precipitation produced by these processes is abundant and plays an important role both in the global land–ocean water circulation (Ogino et al. 2017). On the basis of the findings of previous studies (Yang and Smith 2006; Smith et al. 2012; Yamanaka 2016; Shige et al. 2017), precipitation in coastal areas can be categorized broadly into two major types: that caused by thermal forcing and that caused by mechanical forcing.
Thermally forced precipitation is precipitation driven by differential surface heating between the land and the sea, with diurnal variation in precipitation rate, i.e., a maximum rate in the afternoon (early morning) over coastal land (ocean) areas. This type of precipitation has been widely observed along global tropical coasts (Kikuchi and Wang 2008; Bergemann et al. 2015), and is particularly prominent around the Indonesian Maritime Continent (IMC) (Mori et al. 2004), in northwestern South America (Mapes et al. 2003a), and in the Bay of Bengal (Yang and Slingo 2001). As time progresses, the precipitation peak over the coastal land (ocean) area often propagates in the inland (offshore) direction away from the coastline. Earlier studies attributed the reason for this to the propagation of gravity waves excited by the heating of land surface or elevated terrain (Yang and Slingo 2001; Mapes et al. 2003b; Li and Carbone 2015), the progression of sea–land-breeze fronts as gravity currents (Houze et al. 1981; Bai et al. 2021), advection by mid- to upper-level winds (Yanase et al. 2017), and the propagation of squall lines on the downwind side of mountains (Satomura 2000).
Mechanically forced precipitation occurs when large-scale low-level winds, such as monsoon flows, encounter coastal mountains and are forced to rise, resulting in large amounts of precipitation. This type of precipitation does not have an inherent diurnal cycle provided that the large-scale winds themselves and other environmental variables do not exhibit diurnal variation. This precipitation type prevails on the upwind side of the Asian monsoon flow, such as on the southwestern coast of the Indian Peninsula and the Indochina Peninsula in summer (Shige et al. 2017), and on the eastern coast of Vietnam and the Philippines in winter (Chang et al. 2005), as well as around coastal mountains during periods of strong winds in the active phase of intraseasonal oscillation such as the MJO and the boreal summer monsoon intraseasonal oscillation (Ichikawa and Yasunari 2008; Shige et al. 2017).
Coastal precipitation behaves in a complex manner owing to interplay between thermal and mechanical forcing, and the effects of the two components cannot be distinguished completely. However, previous studies have suggested that mechanical forcing is dominant under the condition of strong low-level background winds, whereas thermal forcing is dominant under calm conditions, as might be expected intuitively (Ichikawa and Yasunari 2008; Nugent et al. 2014; Shige et al. 2017). Reflecting this difference in mechanism, the diurnal variations of precipitation also differ depending on the low-level winds. For example, Shige et al. (2017) classified precipitation events identified from satellite-based observations according to the strength of the coast-normal wind along the southwestern coast of the Indian and Indochina peninsulas, which are on the continental windward side of the Asian summer monsoon flow. They revealed that the mechanically forced precipitation with small diurnal variation is dominant when background winds are strong, while the thermally forced diurnal cycle appears under calm conditions.
However, the effect of thermal forcing does not disappear even in strong winds, and the behavior of the diurnal cycle of precipitation changes depending on the direction and strength of the wind (Wang and Sobel 2017; Peatman et al. 2021; Natoli and Maloney 2022; Yokoi et al. 2022). The sea–land-breeze circulation, the best example of thermal forcing, can be interpreted as the propagation of gravity waves excited by the land–sea surface heating difference (Rotunno 1983). Previous studies in southern China and coastal areas of the IMC have reported that the propagation characteristics of such gravity waves depend on the strength of the background wind, which in turn affects the diurnal cycle of precipitation (Du and Rotunno 2018, hereafter DR18; Short et al. 2019).
Most previous studies on the relationship between the diurnal cycle of precipitation and the background wind field focused on specific regions in specific seasons. Consequently, there is still no consensus regarding common features of such precipitation in different regions, or the relative importance of factors that cause background wind dependence in the diurnal cycle. Recently, Fang and Du (2022) investigated how offshore propagation of precipitation in the global coastal zone varies with the background wind field. However, consideration of precipitation associated with onshore propagation and mechanical forcing was beyond the scope of their analysis. Moreover, they have used the combined products of infrared (IR) and passive microwave (PMW) radiometers, which do not necessarily capture the precise diurnal cycle of precipitation (Hayden and Liu 2021).
The combined products estimate precipitation for areas not observed by PMW with IR observations, assuming that cold clouds with low IR brightness temperatures bring intense precipitation (TRMM 3B42) or that the PMW-derived precipitation area moves along the direction of IR-derived cloud-top movement (CMORPH, GSMaP). In tropical mesoscale convective systems, cold anvil clouds decoupled from the deep convection often become expanded by the upper-level winds in a direction completely different from the propagation direction of the surface heavy rainfall, leading to IR estimating the rainfall area in the incorrect location where these anvil clouds extend (Rickenbach 1999). Furthermore, even in PMW radiometers, the precipitation estimation algorithms essentially differ between land and sea, posing challenges in judging the existence of precipitation in coastal areas where the PMW footprint (approximately 50 km) includes both ocean and land (Mega and Shige 2016). In addition, the PMW algorithm over land conspicuously underestimates shallow but intense rainfall caused by orographically enhanced warm rain processes (Shige et al. 2013).
Meanwhile, spaceborne precipitation radar can provide observations of the diurnal cycle of precipitation that are more consistent with ground-based observations than those derived by other satellite sensors. It is an active sensor that can observe a more realistic precipitation distribution that accounts for the vertical structure and is independent of surface conditions (Nakamura 2021). However, the observations from a spaceborne radar are limited to snapshots of areas with a swath less than 250 km wide along the satellite’s orbit. Because of the narrow observation area and the difficulty of handling this orbital format data, most previous studies on global coastal diurnal cycles have utilized the PMW–IR combined products published as more manageable gridded format data.
In this study, we investigate the relationship between coastal precipitation and low-level winds using a spaceborne radar to gain insight into the globally common determinant mechanism of the diurnal cycle of coastal precipitation. To overcome the limited samplings of the spaceborne radar, we propose a method to effectively use it for coastal precipitation analysis by taking climatological averages based on distance from the coastline (DFC), which is often used as a reference parameter in the study of coastal precipitation (Ogino et al. 2016; Hirose et al. 2017).
The remainder of the paper is organized as follows. Section 2 describes the satellite data and basic parameters used in the analysis, and then it introduces the classification method based on the low-level background wind field. Section 3 presents the classification results for the observational analysis. Section 4 discusses why wind speed dependence is evident in the diurnal cycle of coastal precipitation. Finally, the study findings are summarized in section 5.
2. Data and methods
This study focuses on the diurnal cycle of tropical coastal precipitation, and all subsequent analyses are performed across all longitudes in the tropical region from 22.5°S to 22.5°N. We designated this analysis area based on Ogino et al. (2017)’s definition of the tropics which considers the global zonal mean precipitation and meridional water vapor transport.
a. TRMM PR
We primarily used precipitation data derived from the TRMM (Kummerow et al. 1998) PR (Kozu et al. 2001) level 2 version 06A product (Seto et al. 2021) from January 1998 to December 2014 (17 years), which derive from a Ku-band radar observing the global tropics and subtropics. The PR swath is 215 km wide (245 km after the change in satellite altitude in August 2001) and the horizontal resolution is approximately 5 km. Because the TRMM satellite travels in a solar asynchronous orbit, long-term averaging operation allows the number of observations to be independent of local time. Although the GPM (Hou et al. 2014) DPR (Kojima et al. 2012) is currently operational, we used the predecessor, the PR, for our analysis because the DPR has an observational coverage extending into mid–high latitudes and has fewer samples in the tropics compared to the PR. While the DPR has a higher sensitivity than the PR and is better at detecting light rainfall in subtropics (Hamada and Takayabu 2016), its impact on our analysis is minimal because the contribution of light rain to the total precipitation in the tropics is small (Tang et al. 2017).
b. ERA5
From ERA5 (Hersbach et al. 2020), we used the daily mean 850-hPa wind, vertical velocity profiles, and annual mean water vapor mixing ratio profiles from pressure levels ranging from 1000 to 700 hPa. The original data were at the 1-h temporal resolution on a 0.25° spatial grid. The daily mean 850-hPa wind was calculated as the average of 24 data points per day for each grid point, from 0000 to 2300 UTC. For the case where the 850-hPa height was lower than the surface height, the wind vector 10 m above the ground was used instead. The original vertical velocity data from ERA5 is given as pressure velocity ω. To align the unit with the results of the linear theory described in section 4b, we scaled ω to w in height coordinates by w = −ω/ρg, where g is the gravitational acceleration (9.81 m s−2) and ρ is the density of the atmosphere. ρ was calculated using the gas equation of state as ρ = p/RT, where p is pressure, R is the gas constant (287 J kg−1 K−1), and T is temperature. This scaling presupposes a hydrostatic condition, also an assumption made in the comparative linear theory, thereby posing no issues.
c. Distance from the coastline
The DFC was calculated for each 0.25° × 0.25° cell (shading in Fig. 1a) and the datasets were composited accordingly following Ogino et al. (2016). We used the Global Land One-km Base Elevation topographic dataset (Hastings et al. 1999) with a grid interval of 30 arc s for sea/land determination. Because we focused on precipitation along the coastlines with sufficiently long horizontal extents, we defined land areas smaller than 15 000 km2 as marine areas. Even for small tropical islands that fitted the land exclusion defined above, diurnal variations in precipitation dependent on the sea–land distribution and topography have been reported in Sobel et al. (2011); however, consideration of such phenomena were beyond the scope of this study. The DFC was taken as positive (negative) in the inland (offshore) direction.
Maps of (a) distance from coastline (shading) and cross-shore direction (vectors), and cross-shore wind (shading) and ERA5 daily mean 850-hPa horizontal wind (vectors) on (b) a day in boreal summer (10 Jul 2000) and (c) a day in boreal winter (7 Feb 2001). In (a) and in (b) and (c), the latitude and longitude intervals of the arrows are 2.5° and 4°, respectively. In (b) and (c), shades indicate within 1000 km of the coastline.
Citation: Journal of Climate 37, 1; 10.1175/JCLI-D-23-0180.1
d. Cross-shore wind
To reveal how precipitation characteristics differ depending on the environmental conditions of the low-level wind, the TRMM PR observations were classified according to the ERA5 daily mean 850-hPa horizontal wind in the cross-shore direction (U850). Here, the TRMM PR observations are resampled into 0.25° × 0.25° cells to match the grid of the ERA5. This 0.25° grid size is sufficiently larger than the 5 km footprint of the PR. Following Aoki and Shige (2021), the cross-shore direction was defined as the direction of the nearest coastline for each 0.25° × 0.25° cell, which was then smoothed horizontally using a Gaussian filter with standard deviation of 1° such that the direction is perpendicular to the large-scale coastline (vectors in Fig. 1a). Note that a positive U850 value means from the ocean toward the land (i.e., the onshore direction), and a negative U850 value means from the land toward the ocean (i.e., the offshore direction).
Examples of U850 calculated on a day in boreal summer and on a day in boreal winter are shown in Figs. 1b and 1c, respectively. On a boreal summer day (Fig. 1b) during the summer monsoon season, a westerly wind is predominant in the region of 10°–20°N, which determines that western (eastern) parts of the Indian Peninsula, Indochina Peninsula, and the Philippines are areas with onshore (offshore) wind. In contrast, on a boreal winter day (Fig. 1c) under the easterly winter monsoon winds, western (eastern) parts of the above regions are classified as areas with offshore (onshore) wind. Around the IMC, absolute wind speeds are low in summer, indicating regions with weak winds, whereas wind speeds are strong in winter when the MJO is in its active phase [i.e., phases 4 and 5 following Kikuchi (2020)].
e. Classification methods
Here, we briefly describe the procedure for cross-shore wind classification (see appendix for a detailed description). First, all orbits of the TRMM PR observations were classified and resampled according to U850 (interval: 0.2 m s−1), the DFC (interval: 25 km), and local time (interval: 1 h) in each cell. Because the sampling number of radar observations is larger at higher latitudes owing to the orbital characteristics of the TRMM satellite, the observations were standardized by the total number of observations in each cell according to Eq. (A1).
The 2D histogram in Fig. 2a shows daily averages of the sampling area (sPDF; appendix) with the DFC on the x axis and U850 on the y axis together with the TRMM PR observations across all longitudes in tropical zonal areas (22.5°S–22.5°N). In regions where small islands are clustered together, such as around the IMC, there is no area more than a few hundred kilometers from the coastline; therefore, as the area becomes closer to the coast, the number of samplings increases. Figure 2b shows the mean precipitation rate in the 2D histogram of the DFC and U850. Over the ocean with U850 > 0, the closer the area is to the coastline and the stronger the wind blows in the onshore direction, the more precipitation occurs. This is attributable to the large amount of water vapor evaporated from the ocean and transported toward the coastal area. In the case of offshore background winds (U850 < 0), precipitation in the coastal area tends to diminish as wind speed increases. This is probably because the inflow of dry air from the landward side prevails. In inland areas, precipitation is larger when winds are weak.
(a) Two-dimensional histograms of daily averages of the sampling area (sPDF) and (b) daily mean precipitation rate with DFC on the x axis and U850 on the y axis computed by statistics of TRMM observations in tropical zonal areas (22.5°S–22.5°N). The contours in 2D histograms indicate the classification threshold (Uth) of the 10th, 30th, 70th, and 90th percentiles. The one-dimensional histograms were extracted from the 2D histograms where DFC = −25 to 0 km, colored according to Uth.
Citation: Journal of Climate 37, 1; 10.1175/JCLI-D-23-0180.1
To ensure sufficient samples to capture the diurnal variation, the aggregated data in a 2D histogram were categorized into five classes (contours in Fig. 2). Because wind speeds are systematically weaker in coastal and land areas than over the open ocean owing to the friction of the land surface, simple categorization by wind speed would result in statistics for the ocean and land areas under differing background environments. Thus, the threshold used for classification was not just wind speed but also the percentile value of the number of observations ranked by U850 from negative to positive at each DFC (Uth; see appendix). In the remainder of this paper, Uth in the range of the 0th–10th, 10th–30th, 30th–70th, 70th–90th, and 90th–100th percentiles is referred to as a “Strong Offshore,” “Moderate Offshore,” “Weak,” “Moderate Onshore,” and “Strong Onshore” regime, respectively.
3. Results
a. Geographical distribution
Using the classification method introduced in the previous section, we explore how the diurnal cycle of precipitation changes depending on the cross-shore wind. Figure 3 illustrates the geographic distribution of the cumulative annual amount of TRMM PR precipitation attributable to each regime, along with elevation. The larger values in Figs. 3b–d indicate areas with a higher frequency of days classified under the respective regimes and a greater amount of precipitation during those regimes. According to Figs. 3b–d, each regime predominantly contributes to precipitation in geographically distinct locations. Note that Fig. 3c represents the accumulated precipitation from the events comprising 40% (30th–70th percentiles) of the whole event, whereas Figs. 3b and 3d show the accumulated precipitation from the events comprising 10% (90th–100th and 0th–10th percentiles, respectively).
Maps of (a) topographic elevation and cumulative annual precipitation during (b) Strong Onshore, (c) Weak, and (d) Strong Offshore wind regimes. Areas enclosed by black lines with letters A–H in (a) show the nine geographic regions for which regional analysis was performed in section 3d.
Citation: Journal of Climate 37, 1; 10.1175/JCLI-D-23-0180.1
Cumulative precipitation during the Strong Onshore regime is large in higher latitudes in the tropics, corresponding to the areas where the monsoon flow encounters coastal mountains and mechanical forcing prevails, such as southwestern coastal areas of the Indian and Indochina peninsulas (Shige et al. 2017), eastern and western coasts of the Philippines (Chang et al. 2005), the east coast of Madagascar (Figs. 3a,b). Conversely, precipitation attributed to the Weak regime prevails in tropical low-latitude areas (Fig. 3c), corresponding to the regions where previous studies (Yamanaka 2016; Mapes et al. 2003a) have reported a predominance of thermally forced precipitation with diurnal propagation. Precipitation under the Strong Offshore regime prevails over the downwind side of the continents and peninsulas in the Asian monsoon region, although the amount is small (Fig. 3d).
b. Daily mean and mechanical forcing
Figure 4a presents Hovmöller diagrams showing the diurnal cycle of precipitation rate as observed by the TRMM PR for each wind regime. Before discussing diurnal variations, we first focus on the underlying daily mean precipitation (Fig. 5a).
Time–distance Hovmöller diagrams of (a) TRMM PR precipitation rate, (b) TRMM PR precipitation rate deviations from daily mean, and (c) ERA5 vertical velocity deviations at 900 hPa. For (c), the second mode of the FFT is subtracted to remove the semidiurnal upward and downward motions due to atmospheric tides. A distance of 0 represents the coast, with positive (negative) distance values indicating locations over land (ocean). Local time when the component of the first mode of the FFT reaches its maximum (minimum) is indicated by the solid (dashed) line.
Citation: Journal of Climate 37, 1; 10.1175/JCLI-D-23-0180.1
(a) Daily mean and amplitudes of (b) the first mode and (c) the second and higher modes of the FFT for TRMM PR precipitation rate for each wind regime, plotted as a function of distance from the coastline. (d) The ratio of the amplitude of the first mode to the daily mean, which measures the contribution from the first mode, or thermal forcing.
Citation: Journal of Climate 37, 1; 10.1175/JCLI-D-23-0180.1
The daily mean precipitation rate near the coastline increases as onshore winds get stronger. This is because evaporation from the ocean and water vapor transport toward coastal areas are greater under stronger onshore background wind conditions, as discussed in section 2. In the Strong Onshore regime, precipitation gradually increases from a few hundred kilometers offshore to nearer the coastline. This increase is because of the barrier of coastal mountains and the rougher surface of the land surface in comparison with that of the sea, which causes the low-level winds to slow and converge in front of the land. Conversely, stronger offshore winds reduce the daily mean precipitation, indicating that advection of continental dry air results in less precipitation. These deductions are consistent with the geographic distribution of precipitation contributed by these regimes presented in Fig. 3. The mean precipitation under the Weak regime is less than the onshore regimes and is rather comparable to that in the offshore regimes.
c. Diurnal variations and thermal forcing
Figure 4b presents Hovmöller diagrams showing the precipitation deviations from the daily mean. The well-known characteristic of thermally forced precipitation around tropical coasts, with a morning peak over the sea and an afternoon peak over land, is evident for all wind regimes. Both peaks separately propagate away from the coastline forming “envelopes” of diurnal precipitation anomalies in the Hovmöller diagrams (Fig. 4b). This characteristic corresponds to the offshore propagation commonly observed in the IMC (Mori et al. 2004), northwestern South America (Mapes et al. 2003a), and the Bay of Bengal (Yang and Slingo 2001) and the inland propagation observed over the Indo-China Peninsula (Satomura 2000) and the northeastern coast of Brazil (Garreaud and Wallace 1997).
Of interest here is the difference in the diurnal variations of precipitation depending on wind regime. To compare the characteristics of the diurnal cycle among the five wind regimes, we used the first Fourier transform (FFT) to calculate the amplitude and phase of the first mode (hereafter diurnal amplitude and diurnal phase) from the time series at each DFC (Fig. 5b and lines in Fig. 4b). Over coastal land (up to approximately 50 km inland from the coast), the amplitudes of the second and higher frequency modes of FFT are comparable to that of the first mode (Figs. 5b,c) because diurnal variations cannot be expressed by a sinusoidal function with a diurnal period (Oki and Musiake 1994). However, the time of maximum precipitation broadly coincides with the peak of the first mode (Fig. 4b); therefore, we focus only on the first mode.
In each regime, the diurnal amplitude exhibits two peaks, each corresponding to the propagation envelopes over the ocean and the land (Fig. 5b). The amplitude tends to be larger in the Weak regime and smaller in the strong wind regimes. This could be due to the enhanced mechanical turbulent mixing under strong wind relaxing the surface heating contrast (Nugent et al. 2014). In addition, the Weak regime incorporates a larger proportion of data samples from near the equator, where solar radiation is stronger, resulting in more intense diurnal variation than in the stronger wind regimes (Fig. 3c).
The diurnal amplitude serves as a measure of thermally forced precipitation. Near the coastline and over coastal ocean (−150 to 50 km), the ratio of the diurnal amplitude to the daily mean precipitation is approximately one-half in the Weak regime (Fig. 5d). However, it diminishes to roughly one-fourth in the Strong Onshore regime. This indicates that the contribution of mechanically forced precipitation to total precipitation is larger as the wind gets stronger in the onshore direction, consistent with previous studies in some tropical regions that have suggested that diurnal variations are weaker when strong winds blow from the ocean (Shige et al. 2017; Peatman et al. 2021; Natoli and Maloney 2022).
Yet, the diurnal variation cannot be ignored even under strong background winds (Fig. 4b), which is consistent with past studies based on satellite observations in the Bay of Bengal (Shige et al. 2017) and the IMC (Yanase et al. 2017; Peatman et al. 2021), and numerical simulations (Wang and Sobel 2017). In stronger wind regimes, both peaks of diurnal amplitude appear at locations further downwind than those in the Weak regime, i.e., the diurnal envelope on the upwind side of the coastline is more horizontally squashed, while that on the downwind side is more horizontally stretched. Near the coastline, the diurnal amplitude has a local minimum between these two peaks, where the sign switches between the seaward and landward sides. In stronger wind regimes, the minimum becomes less noticeable than in the Weak regime.
A major difference in the diurnal phase among the regimes is evident in Fig. 4b. In the Weak regime, the propagation pattern is symmetrical with opposite signs across the coastline. However, in stronger wind regimes, the pattern becomes asymmetric owing to shift in the time of maximum and minimum precipitation. Figure 6a illustrates the lines of the diurnal phase peaks and minimums for each wind regime (Fig. 4b) in the same figure. In comparison with the Weak regime, the time of the peak in the diurnal phase occurs earlier on the downwind side of the background wind, i.e., on the land side in the Strong Onshore regime and on the ocean side in the Strong Offshore regime (Fig. 6a). In stronger wind regimes, the diurnal phase difference is particularly pronounced within 100 km downwind of the coast with slow propagation speed. Additionally, on the upwind side of the background wind, i.e., on the land side in the Strong Offshore regime and on the ocean side in the Strong Onshore regime, the diurnal phase is delayed, contrary to the case on the downwind side.
Time–distance diagram of the phases of the first mode of the FFT for (a) precipitation rate and (b) vertical velocity at 900 hPa in each of the wind regimes at each distance from the coastline. Local time when the component of the diurnal frequency mode of the FFT reaches its maximum (minimum) is indicated by the solid (dashed) line.
Citation: Journal of Climate 37, 1; 10.1175/JCLI-D-23-0180.1
d. Regional analysis
We further performed the same analysis as in sections 3b and 3c for the nine segmented regions shown in Fig. 3a, ensuring that the relationship between the diurnal cycle of precipitation and the background wind is coherent for all tropical regions. Figure 7 presents Hovmöller diagrams for each wind regime in respective regions. Here, the same classification thresholds were used as those used for the entire region of the tropics. Although the precipitation rate varies greatly from region to region, a common pattern of diurnal precipitation propagation can be observed for each regime, regardless of the location. Figure 8 shows the results of the FFT decomposition: the daily mean (Fig. 8a), diurnal amplitude (Fig. 8b), and diurnal phase (Fig. 8c). In all regions, as the onshore winds become stronger, the daily mean values along the coast increase, consistent with the case of the entire tropics. The diurnal phase in each region shows similar phase delay and phase advance according to the background wind as observed for the entire tropics, except for northwestern South America where the sample of precipitation is limited.
As in Fig. 4a, but for TRMM PR precipitation rate in the nine regions shown in Fig. 3a. Blank indicates pixels with sPDF of <5000 km2.
Citation: Journal of Climate 37, 1; 10.1175/JCLI-D-23-0180.1
(a) As in Fig. 5a, (b) as in Fig. 5b, and (c) as in Fig. 6a, but for TRMM PR precipitation rate in the nine regions shown in Fig. 3a. Blank indicates pixels with sPDF of <5000 km2.
Citation: Journal of Climate 37, 1; 10.1175/JCLI-D-23-0180.1
4. Discussion
a. Vertical velocity perturbation in ERA5
In this section, we discuss the reason why propagation patterns shown in Fig. 4b depend on the background winds. Figure 4c presents Hovmöller diagrams for the ERA5 900-hPa vertical velocity deviations (
Figure 9 shows the vertical profile of the vertical velocity deviation (w′) under each regime. In the Weak regime (Fig. 9b), afternoon land heating triggers sea breeze anomalies, causing upward motion on the land side of the coast, and corresponding downward motion on the ocean side. Conversely, morning land cooling causes land breeze anomalies, with upward motion over the sea and downward motion over land. Moreover, these figures show the diagonally aligned phase lines, with a symmetrical pattern of opposite signs across the coastline (lines in Fig. 9b). Such phase patterns with oblique tuned ray paths might be the result of coastal gravity waves excited by the diurnally oscillating surface heating contrast between the sea and the land (Rotunno 1983).
Distance–height diagrams of the vertical velocity deviations over a diurnal cycle at 2200, 0100, 0400, 0700, 1000, 1300, 1600, and 1900 LT from the 17-yr-mean ERA5 data for (a) Strong Onshore, (b) Weak, and (c) Strong Offshore regimes. Shading in warm (cold) colors indicates upward (downward) motion. The second mode of the FFT is subtracted to remove the semidiurnal upwelling and downwelling due to atmospheric tides. Solid (dashed) lines are the lines connecting the points where the upward motion reaches its local maximum (minimum). Black (white) lines correspond to the I1 and I2 modes (I3 mode) proposed by Qian et al. (2009).
Citation: Journal of Climate 37, 1; 10.1175/JCLI-D-23-0180.1
In the Strong Onshore regime (Fig. 9a) and the Strong Offshore regime (Fig. 9c), the pattern of w′ is asymmetric across the coastline. Under the Strong Onshore regime, the ray paths are more steeply inclined to the vertical over the ocean and less steeply inclined over the land (black lines). Additionally, a pattern with a negative ray path slope appears within 100 km landward of the coastline (white lines), although it is not as distinct as the black phase lines. In contrast, under the Strong Offshore regime, the ray path slope is more tilted toward the seaward side (black lines), and a pattern with a positive ray path slope appears within 100 km seaward of the coastline (white lines). The vertical propagation patterns (Fig. 9) are similar to those of coastal gravity waves under background winds proposed by Qian et al. (2009; hereafter, Q09). DR18 further suggested that coastal gravity waves form an asymmetric pattern of diurnal precipitation propagation in the onshore background case of the southern coast of China during summer. Therefore, the asymmetry in Fig. 4b appears caused by coastal gravity waves under the background wind.
b. Linear theory of coastal gravity waves
Gravity waves caused by ocean–land surface heating differences have attracted attention as a possible cause of the diurnal propagation of precipitation in observations, numerical simulations, and theoretical studies (Yang and Slingo 2001; Li and Carbone 2015; DR18), though most of them focused only on offshore propagation. They can be described by linear equations, and an exact solution can be obtained from the linear theory of the sea–land-breeze circulation, given a simplified diurnally oscillating heating source over the land (Rotunno 1983; Q09). Here, we examine whether gravity waves alone can explain the diurnal variations in coastal precipitation using the linear theory under background wind proposed by Q09.
Figure 10 shows Hovmöller diagrams for each mode of w′ at an altitude of 1 km in each regime calculated by using the solution of Q09. In our classification, the background wind speed can have various values, even within the same wind regime (Fig. 2a). Therefore, we took the ensemble mean of w′ across various values of U for each regime. Specifically, we performed linear calculations while varying U from −15 to 15 m s−1 in increments of 0.2 m s−1, then we averaged the w′ results from each calculation, weighted by the number of observations for each U850 in the coastal area (the 1D histogram in Fig. 2a). The solution of w′ for the case U < 0 is the reflection about the x = 0 axis with the sign reversed. The parameters employed in Q09 were set at Q0 = 0.5 × 10−5 m s−3, L = 10 km, H = 1 km, and static stability N = 0.01 s−1, which are typical values in tropics (in the online supplemental material). In Q09, the equatorial case is considered, and Coriolis parameter is set to zero. DR18 expanded on Q09 by adding the Coriolis term at a latitude of 20°N (near the edge of our analysis area) into their linear equations; however, the primary behavior of coastal gravity waves does not change significantly.
Time–distance Hovmöller diagrams of the vertical velocity perturbation due to (a) the sum of the I1, I2, and I3 modes; (b) the sum of the I1 and I2 modes; (c) seaward-propagating component of the I1 and I2 modes; (d) landward-propagating component of the I1 and I2 modes; and (e) the I3 mode at 1-km height calculated using the solutions of the linear theory by Qian et al. (2009). The diagram of the Weak regime in (c) includes the I1 mode under onshore background winds and the I2 mode under offshore background winds, and vice versa in (d). The ensemble mean for each wind regime is shown. Phase lines are not shown for amplitude of <0.005.
Citation: Journal of Climate 37, 1; 10.1175/JCLI-D-23-0180.1
Figure 10a, which sums up all three modes, shows the symmetrical pattern during weak winds and the asymmetrical pattern during strong winds, similar to that found in TRMM PR and ERA5. Furthermore, the amplitude of w′ in the linear model is of the same order of magnitude as that in ERA5 (Fig. 4c), indicating that the climatological mean coastal diurnal cycle of w′ can be represented by considering solely the sea–land heating contrast.
The vertical profiles of w′ in Fig. 11 show that the I1 and I2 modes have ray paths that radiate away from the surface at the coastline. When the wind is weak, the ray paths of the I1 and I2 modes are symmetrical across the coastline. However, as the background wind strengthens, the ray paths of the I1 (I2) mode on the upwind (downwind) side of the coastline are more (less) steeply inclined to the vertical owing to Doppler shifting. This is consistent with the characteristics of the phase line inclinations in the ERA5 composite identified in Fig. 9. Consequently, on the upwind side of the coastline, the propagation speed of the I1 mode during stronger wind regimes is slower than during the Weak regime (Figs. 10c,d), leading to a later diurnal phase and a narrower horizontal extent (Fig. 10b). Conversely, on the downwind side of the coastline, the I2 mode during stronger wind regimes propagates faster than during the Weak regime (Figs. 10c,d), resulting in an earlier diurnal phase and a broader horizontal extent (Fig. 10b). However, the earlier diurnal phase of the I2 mode appears only at a distance of >100 km downwind of the coast because the I2 mode is advected by background winds and shifts away from the coastline.
Distance–height diagrams of the vertical velocity deviations due to (a) the sum of the I1, I2, and I3 modes; (b) the sum of the I1 and I2 modes; and (e) the I3 mode at 0600 LT from the linear theory by Qian et al. (2009). The ensemble mean for each wind regime is shown. Dashed lines for (b) show the ray paths for the U = 6, 3, 0, −3, and −6 m s−1 solution of the I1 and I2 modes.
Citation: Journal of Climate 37, 1; 10.1175/JCLI-D-23-0180.1
The I3 mode has large amplitude within approximately 100 km of the coastline, and the amplitude increases as the background wind strengthens (Fig. 10e). The vertical propagation pattern of the I3 mode is close to that of a flow past a steady heat source (Lin and Smith 1986), with a negative (positive) slope in the onshore (offshore) regimes (Fig. 11c), which is also identified in the ERA5 composite near the coast (Fig. 9). In a frame moving at the same speed as the background wind, both I1 and I3 modes embedded within the same wave beam that propagates toward the upwind side (Fig. 11a), while I2 propagates toward the downwind side. Among the waves propagating upwind, the I3 mode is the component that does not have a speed fast enough to propagate upwind against the background wind and is advected downwind relative to the coastline (Fig. 11c). Therefore, in the Strong Onshore (Offshore) regime, the I3 mode has a maximum upward (downward) motion at the coastline at approximately 0300 LT (Fig. 10e), which is the same time as the I1 mode does. Consequently, in stronger wind regimes, the I3 mode causes an earlier phase with slower propagation speed appears near the coastline (Fig. 10a).
Vertical profiles of mean mixing ratio and its vertical derivative used in the calculation of
Citation: Journal of Climate 37, 1; 10.1175/JCLI-D-23-0180.1
(a)–(c) As in Fig. 5b and (d)–(f) as in Fig. 6, but for
Citation: Journal of Climate 37, 1; 10.1175/JCLI-D-23-0180.1
Although our linear calculations impose identical diabatic heating and water vapor mixing ratios for all regimes, these values can vary by location and season. While the
5. Conclusions
No strong consensus exists on the effect of the background wind on the mechanism of the diurnal cycle of coastal precipitation across the tropics. Using spaceborne radar data, this study represented the first global investigation of how the diurnal cycle of coastal precipitation is controlled by the background wind field. We classified the 17-yr observations of the TRMM PR according to the cross-shore component of the low-level wind, demonstrating that the diurnal cycle of coastal precipitation varies depending on the direction (onshore or offshore) and strength of the wind (Fig. 4a).
The daily mean precipitation rate increases as the background wind strengthens in the onshore direction (Fig. 5a), indicating that a large amount of water vapor evaporated from the ocean and transported toward the coastal area, leading to a more significant contribution of mechanically forced precipitation during the Strong Onshore regime. This regime is dominant over windward land areas in monsoon flows. In contrast, precipitation under the Strong Offshore regime prevails over the downwind side of the continents and peninsulas, although the amount is small. The mean precipitation under the Weak regime is less than the onshore regimes and is rather comparable to that in the offshore regimes.
The precipitation deviation from the daily mean is regarded as the component of thermally forced precipitation. In the Weak regime, the ratio of the diurnal amplitude of precipitation to the daily mean is larger than that in the Strong Onshore regime, indicating a more significant contribution of thermal forcing (Fig. 5d). However, the effect of the diurnal cycle remains nonnegligible even in stronger wind regimes. In each regime, the Hovmöller diagrams showing precipitation deviations indicate a morning maximum over the ocean and an afternoon maximum over land, both propagating away from the coastline. The propagation pattern varies depending on the background wind (Fig. 4b). In the Weak regime, the phase pattern of precipitation is symmetric with opposite signs. However, in stronger wind regimes, the phase pattern becomes asymmetric, with the later phase on the upwind side and the earlier phase on the downwind side (Fig. 6a).
Using the linear theory of sea–land breeze circulation under background winds proposed by Q09, we demonstrated that the diurnal precipitation propagation for each wind regime is attributable to the coastal gravity waves excited by land–ocean heating difference (Fig. 13). Without of background wind, the two symmetrical modes of landward and seaward gravity waves result in symmetrical propagation. However, in the stronger wind regimes, the two modes are Doppler-shifted, with the mode on the upwind side (I1) having a slower propagation speed and the mode on the downwind side (I2) having a faster propagation speed. This results in a delayed diurnal phase on the upwind side and an advanced diurnal phase beyond 100 km on the downwind side. Furthermore, gravity waves propagating toward the upwind side in the flow-relative frame but with slower group velocity are advected to the downwind side, forming the I3 mode that moves in the downwind direction more slowly than the background wind. This mode has a maximum upward motion at the coastline at the same time as the I1 mode does (e.g., 0300 LT in the Strong Onshore regime). Consequently, during stronger wind regimes, this mode causes an earlier diurnal phase on the downwind near the coast than in the calm conditions, forming asymmetric propagation patterns.
In addition, coastal gravity waves can account for the differences in the diurnal amplitude of TRMM PR precipitation among wind regimes. Under strong wind conditions, the two peaks in diurnal amplitude appear further downwind than those under weak wind conditions owing to the Doppler shift of the I1 and I2 modes. Moreover, the local minimum between these two peaks becomes less noticeable during strong winds owing to the presence of the I3 mode.
While the diurnal cycle of coastal precipitation is well-explained by gravity waves arising from the land–ocean heating difference in the climatological context, the role of these waves in individual precipitation events remains unaddressed. The intensity of vertical velocity presented in this study is, at most, a few centimeters per second. It is difficult to presume that an updraft of this magnitude could directly generate new precipitating clouds in areas with unsaturated clear air conditions. Instead, as proposed in previous studies (Love et al. 2011; Hassim et al. 2016; Yokoi et al. 2017), gravity waves likely contribute to precipitation by creating conducive conditions by increasing moisture through vertical water vapor transport and by destabilizing the atmosphere by cooling the environmental virtual temperature. In individual events, convection can be triggered by local phenomena that produce stronger updrafts, such as frontogenesis of density currents, or gravity waves related to land–sea breezes or diabatic heating from deep convection (Shige and Satomura 2000, 2001; Fovell 2005; Bai et al. 2021). However, these localized phenomena have a small horizontal scale (within tens of kilometers), a short lifespan, and considerable variations in their occurrence locations. Conversely, the sea–land heating contrast recurring on a diurnal cycle provide a steady forcing regardless of location and season, and they operate on a larger scale. As a result, the response to the sea–land heating contrast is probably more evident in the climatological average. Determining the impact of these local-scale phenomena on climatological precipitation requires additional research using numerical simulations or high-resolution observations, though such investigation is beyond the scope of this study.
Our classification method helps validate global weather and climate models, which are challenged in reproducing the diurnal phase and amplitude of precipitation (Dai and Trenberth 2004; Covey et al. 2016). Previously, evaluation of the diurnal representation of precipitation in models has often been performed by dividing the globe into two regions: the ocean and the land (Dai 2006; Christopoulos and Schneider 2021). However, Ogino et al. (2017) highlighted the importance of incorporating the coastal area as a third region owing to its substantial precipitation when examining the global land–ocean hydrological cycle. This study’s findings reinforce the significance of adopting this “three-region” concept for a more comprehensive understanding of the climatological characteristics of diurnal cycle and for an improved assessment of their representation in models. While we focus on tropical areas, the diurnal cycle at mid- to high latitudes may be influenced by factors like reduced insolation and significant changes in coastal gravity wave behavior around latitude 30° suggested by Rotunno (1983). By utilizing the GPM DPR and extending our analysis to higher latitudes, future research would reveal more universal characteristics of coastal diurnal cycles.
Acknowledgments.
This work was supported by JST, the establishment of university fellowships toward the creation of science technology innovation Grant JPMJFS2123; third Research Announcement on the Earth Observations (EO-RA3) of the Japan Aerospace Exploration Agency (JAXA) (ER3GPF006); JSPS KAKENHI Grants 18KK0130, 19H01969, and 23KJ1286; and a joint research program of CEReS, Chiba University (2020/21). We thank the anonymous reviewers for their insightful and constructive comments and suggestions.
Data availability statement.
TRMM PR data were provided by Japan Aerospace Exploration Agency (JAXA). ERA5 data were obtained through the Climate Data Store (CDS).
APPENDIX
Details of the Classification Procedure
(a) Histogram of the number of observations (sPDF) of the cross-shore wind (m s−1) and (b) mean precipitation rate in each cross-shore wind bin in the 0°, 100°E cell.
Citation: Journal of Climate 37, 1; 10.1175/JCLI-D-23-0180.1
REFERENCES
Aoki, S., and S. Shige, 2021: Large precipitation gradients along the south coast of Alaska revealed by spaceborne radars. J. Meteor. Soc. Japan, 99, 5–25, https://doi.org/10.2151/jmsj.2021-001.
Bai, H., and Coauthors, 2021: Formation of nocturnal offshore rainfall near the west coast of Sumatra: Land breeze or gravity wave? Mon. Wea. Rev., 149, 715–731, https://doi.org/10.1175/MWR-D-20-0179.1.
Bergemann, M., C. Jakob, and T. P. Lane, 2015: Global detection and analysis of coastline-associated rainfall using an objective pattern recognition technique. J. Climate, 28, 7225–7236, https://doi.org/10.1175/JCLI-D-15-0098.1.
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.
Christopoulos, C., and T. Schneider, 2021: Assessing biases and climate implications of the diurnal precipitation cycle in climate models. Geophys. Res. Lett., 48, e2021GL093017, https://doi.org/10.1029/2021GL093017.
Covey, C., P. J. Gleckler, C. Doutriaux, D. N. Williams, A. Dai, J. Fasullo, K. Trenberth, and A. Berg, 2016: Metrics for the diurnal cycle of precipitation: Toward routine benchmarks for climate models. J. Climate, 29, 4461–4471, https://doi.org/10.1175/JCLI-D-15-0664.1.
Dai, A., 2006: Precipitation characteristics in eighteen coupled climate models. J. Climate, 19, 4605–4630, https://doi.org/10.1175/JCLI3884.1.
Dai, A., and K. E. Trenberth, 2004: The diurnal cycle and its depiction in the Community Climate System Model. J. Climate, 17, 930–951, https://doi.org/10.1175/1520-0442(2004)017<0930:TDCAID>2.0.CO;2.
Du, Y., and R. Rotunno, 2018: Diurnal cycle of rainfall and winds near the south coast of China. J. Atmos. Sci., 75, 2065–2082, https://doi.org/10.1175/JAS-D-17-0397.1.
Fang, J., and Y. Du, 2022: A global survey of diurnal offshore propagation of rainfall. Nat. Commun., 13, 7437, https://doi.org/10.1038/s41467-022-34842-0.
Fovell, R. G., 2005: Convective initiation ahead of the sea-breeze front. Mon. Wea. Rev., 133, 264–278, https://doi.org/10.1175/MWR-2852.1.
Garreaud, R. D., and J. M. Wallace, 1997: The diurnal march of convective cloudiness over the Americas. Mon. Wea. Rev., 125, 3157–3171, https://doi.org/10.1175/1520-0493(1997)125<3157:TDMOCC>2.0.CO;2.
Hamada, A., and Y. N. Takayabu, 2016: Improvements in detection of light precipitation with the Global Precipitation Measurement Dual-Frequency Precipitation Radar (GPM DPR). J. Atmos. Oceanic Technol., 33, 653–667, https://doi.org/10.1175/JTECH-D-15-0097.1.
Hassim, M. E. E., T. P. Lane, and W. W. Grabowski, 2016: The diurnal cycle of rainfall over New Guinea in convection-permitting WRF simulations. Atmos. Chem. Phys., 16, 161–175, https://doi.org/10.5194/acp-16-161-2016.
Hastings, D. A., and Coauthors, 1999: The Global Land One-kilometer Base Elevation (GLOBE) digital elevation model, version 1.0. NOAA, National Geophysical Data Center, https://www.ngdc.noaa.gov/mgg/fliers/globedem.html.
Hayden, L., and C. Liu, 2021: Differences in the diurnal variation of precipitation estimated by spaceborne radar, passive microwave radiometer, and IMERG. J. Geophys. Res. Atmos., 126, e2020JD033020, https://doi.org/10.1029/2020JD033020.
Hersbach, H., and Coauthors, 2020: The ERA5 global reanalysis. Quart. J. Roy. Meteor. Soc., 146, 1999–2049, https://doi.org/10.1002/qj.3803.
Hirose, M., Y. N. Takayabu, A. Hamada, S. Shige, and M. K. Yamamoto, 2017: Spatial contrast of geographically induced rainfall observed by TRMM PR. J. Climate, 30, 4165–4184, https://doi.org/10.1175/JCLI-D-16-0442.1.
Hou, A. Y., and Coauthors, 2014: The Global Precipitation Measurement (GPM) mission. Bull. Amer. Meteor. Soc., 95, 701–722, https://doi.org/10.1175/BAMS-D-13-00164.1.
Houze, R. A., Jr., S. G. Geotis, F. D. Marks Jr., and A. K. West, 1981: Winter monsoon convection in the vicinity of North Borneo. Part I: Structure and time variation of the clouds and precipitation. Mon. Wea. Rev., 109, 1595–1614, https://doi.org/10.1175/1520-0493(1981)109<1595:WMCITV>2.0.CO;2.
Ichikawa, H., and T. Yasunari, 2008: Intraseasonal variability in diurnal rainfall over New Guinea and the surrounding oceans during austral summer. J. Climate, 21, 2852–2868, https://doi.org/10.1175/2007JCLI1784.1.
Kikuchi, K., 2020: Extension of the bimodal intraseasonal oscillation index using JRA-55 reanalysis. Climate Dyn., 54, 919–933, https://doi.org/10.1007/s00382-019-05037-z.
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.
Kojima, M., and Coauthors, 2012: Dual-frequency precipitation radar (DPR) development on the global precipitation measurement (GPM) core observatory. Proc. SPIE, 8528, 85281A, https://doi.org/10.1117/12.976823.
Kozu, T., and Coauthors, 2001: Development of precipitation radar onboard the Tropical Rainfall Measuring Mission (TRMM) satellite. IEEE Trans. Geosci. Remote Sens., 39, 102–116, https://doi.org/10.1109/36.898669.
Kummerow, C., W. Barnes, T. Kozu, J. Shiue, and J. Simpson, 1998: The Tropical Rainfall Measuring Mission (TRMM) sensor package. J. Atmos. Oceanic Technol., 15, 809–817, https://doi.org/10.1175/1520-0426(1998)015<0809:TTRMMT>2.0.CO;2.
Li, Y., and R. E. Carbone, 2015: Offshore propagation of coastal precipitation. J. Atmos. Sci., 72, 4553–4568, https://doi.org/10.1175/JAS-D-15-0104.1.
Lin, Y. L., and R. B. Smith, 1986: Transient dynamics of airflow near a local heat source. J. Atmos. Sci., 43, 40–49, https://doi.org/10.1175/1520-0469(1986)043<0040:TDOANA>2.0.CO;2.
Love, B. S., A. J. Matthews, and G. M. S. Lister, 2011: The diurnal cycle of precipitation over the Maritime Continent in a high-resolution atmospheric model. Quart. J. Roy. Meteor. Soc., 137, 934–947, https://doi.org/10.1002/qj.809.
Mapes, B. E., T. T. Warner, M. Xu, and A. J. Negri, 2003a: Diurnal patterns of rainfall in northwestern South America. Part I: Observations and context. Mon. Wea. Rev., 131, 799–812, https://doi.org/10.1175/1520-0493(2003)131<0799:DPORIN>2.0.CO;2.
Mapes, B. E., T. T. Warner, and M. Xu, 2003b: Diurnal patterns of rainfall in northwestern South America. Part III: Diurnal gravity waves and nocturnal convection offshore. Mon. Wea. Rev., 131, 830–844, https://doi.org/10.1175/1520-0493(2003)131<0830:DPORIN>2.0.CO;2.
Mega, T., and S. Shige, 2016: Improvements of rain/no-rain classification methods for microwave radiometer over coasts by dynamic surface-type classification. J. Atmos. Oceanic Technol., 33, 1257–1270, https://doi.org/10.1175/JTECH-D-15-0127.1.
Mori, S., and Coauthors, 2004: Diurnal land–sea rainfall peak migration over Sumatera Island, Indonesian Maritime Continent, observed by TRMM satellite and intensive rawinsonde soundings. Mon. Wea. Rev., 132, 2021–2039, https://doi.org/10.1175/1520-0493(2004)132<2021:DLRPMO>2.0.CO;2.
Nakamura, K., 2021: Progress from TRMM to GPM. J. Meteor. Soc. Japan, 99, 697–729, https://doi.org/10.2151/jmsj.2021-035.
Natoli, M. B., and E. D. Maloney, 2022: The tropical diurnal cycle under varying states of the monsoonal background wind. J. Atmos. Sci., 80, 235–258, https://doi.org/10.1175/JAS-D-22-0045.1.
Nugent, A. D., R. B. Smith, and J. R. Minder, 2014: Wind speed control of tropical orographic convection. J. Atmos. Sci., 71, 2695–2712, https://doi.org/10.1175/JAS-D-13-0399.1.
Ogino, S.-Y., M. D. Yamanaka, S. Mori, and J. Matsumoto, 2016: How much is the precipitation amount over the tropical coastal region? J. Climate, 29, 1231–1236, https://doi.org/10.1175/JCLI-D-15-0484.1.
Ogino, S.-Y., M. D. Yamanaka, S. Mori, and J. Matsumoto, 2017: Tropical coastal dehydrator in global atmospheric water circulation. Geophys. Res. Lett., 44, 11 636–11 643, https://doi.org/10.1002/2017GL075760.
Oki, T., and K. Musiake, 1994: Seasonal change of the diurnal cycle of precipitation over Japan and Malaysia. J. Appl. Meteor., 33, 1445–1463, https://doi.org/10.1175/1520-0450(1994)033<1445:SCOTDC>2.0.CO;2.
Peatman, S. C., J. Schwendike, C. E. Birch, J. H. Marsham, A. J. Matthews, and G.-Y. Yang, 2021: A local-to-large scale view of Maritime Continent rainfall: Control by ENSO, MJO, and equatorial waves. J. Climate, 34, 8933–8953, https://doi.org/10.1175/JCLI-D-21-0263.1.
Qian, T., C. C. Epifanio, and F. Zhang, 2009: Linear theory calculations for the sea breeze in a background wind: The equatorial case. J. Atmos. Sci., 66, 1749–1763, https://doi.org/10.1175/2008JAS2851.1.
Rickenbach, T. M., 1999: Cloud-top evolution of tropical oceanic squall lines from radar reflectivity and infrared satellite data. Mon. Wea. Rev., 127, 2951–2976, https://doi.org/10.1175/1520-0493(1999)127<2951:CTEOTO>2.0.CO;2.
Rotunno, R., 1983: On the linear theory of the land and sea breeze. J. Atmos. Sci., 40, 1999–2009, https://doi.org/10.1175/1520-0469(1983)040<1999:OTLTOT>2.0.CO;2.
Sakazaki, T., and K. Hamilton, 2017: Physical processes controlling the tide in the tropical lower atmosphere investigated using a comprehensive numerical model. J. Atmos. Sci., 74, 2467–2487, https://doi.org/10.1175/JAS-D-17-0080.1.
Satomura, T., 2000: Diurnal variation of precipitation over the Indo-China Peninsula. J. Meteor. Soc. Japan, 78, 461–475, https://doi.org/10.2151/jmsj1965.78.4_461.
Seto, S., T. Iguchi, R. Meneghini, J. Awaka, T. Kubota, T. Maski, and N. Takahashi, 2021: The precipitation rate retrieval algorithms for the GPM dual-frequency precipitation radar. J. Meteor. Soc. Japan, 99, 205–237, https://doi.org/10.2151/jmsj.2021-011.
Shige, S., and T. Satomura, 2000: The gravity wave response in the troposphere around deep convection. J. Meteor. Soc. Japan, 78, 789–801, https://doi.org/10.2151/jmsj1965.78.6_789.
Shige, S., and T. Satomura, 2001: Westward generation of eastward-moving tropical convective bands in TOGA COARE. J. Atmos. Sci., 58, 3724–3740, https://doi.org/10.1175/1520-0469(2001)058<3724:WGOEMT>2.0.CO;2.
Shige, S., S. Kida, H. Ashiwake, T. Kubota, and K. Aonashi, 2013: Improvement of TMI rain retrievals in mountainous areas. J. Appl. Meteor. Climatol., 52, 242–254, https://doi.org/10.1175/JAMC-D-12-074.1.
Shige, S., Y. Nakano, and M. K. Yamamoto, 2017: Role of orography, diurnal cycle, and intraseasonal oscillation in summer monsoon rainfall over the Western Ghats and Myanmar coast. J. Climate, 30, 9365–9381, https://doi.org/10.1175/JCLI-D-16-0858.1.
Short, E., C. L. Vincent, and T. P. Lane, 2019: Diurnal cycle of surface winds in the Maritime Continent observed through satellite scatterometry. Mon. Wea. Rev., 147, 2023–2044, https://doi.org/10.1175/MWR-D-18-0433.1.
Smith, R. B., and Coauthors, 2012: Orographic precipitation in the tropics: The Dominica Experiment. Bull. Amer. Meteor. Soc., 93, 1567–1579, https://doi.org/10.1175/BAMS-D-11-00194.1.
Sobel, A. H., C. D. Burleyson, S. E. Yuter, and M. Biasutti, 2011: Rain on small tropical islands. J. Geophys. Res., 118, 2301–2302, https://doi.org/10.1029/2010JD014695.
Tang, G., Y. Wen, J. Gao, D. Long, Y. Ma, W. Wan, and Y. Hong, 2017: Similarities and differences between three coexisting spaceborne radars in global rainfall and snowfall estimation. Water Resour. Res., 53, 3835–3853, https://doi.org/10.1002/2016WR019961.
Wang, S., and A. H. Sobel, 2017: Factors controlling rain on small tropical islands: Diurnal cycle, large-scale wind speed, and topography. J. Atmos. Sci., 74, 3515–3532, https://doi.org/10.1175/JAS-D-16-0344.1.
Yamanaka, M. D., 2016: Physical climatology of Indonesian Maritime Continent: An outline to comprehend observational studies. Atmos. Res., 178–179, 231–259, https://doi.org/10.1016/j.atmosres.2016.03.017.
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, G.-Y., and J. Slingo, 2001: The diurnal cycle in the tropics. Mon. Wea. Rev., 129, 784–801, https://doi.org/10.1175/1520-0493(2001)129<0784:TDCITT>2.0.CO;2.
Yang, S., and E. A. Smith, 2006: Mechanisms for diurnal variability of global tropical rainfall observed from TRMM. J. Climate, 19, 5190–5226, https://doi.org/10.1175/JCLI3883.1.
Yokoi, S., S. Mori, M. Katsumata, B. Geng, K. Yasunaga, F. Syamsudin, Nurhayati, and K. Yoneyama, 2017: Diurnal cycle of precipitation observed in the western coastal area of Sumatra Island: Offshore preconditioning by gravity waves. Mon. Wea. Rev., 145, 3745–3761, https://doi.org/10.1175/MWR-D-16-0468.1.
Yokoi, S., R. Shirooka, K. Yoneyama, E. O. Cayanan, and C. O. Iglesia, 2022: Diurnal to intraseasonal precipitation variation around the northwestern coast of Luzon Island: Results from YMC-BSM 2018 field campaign. Geophys. Res. Lett., 49, e2022GL098519, https://doi.org/10.1029/2022GL098519.
