Gravity Waves and Other Mechanisms Modulating the Diurnal Precipitation over One of the Rainiest Spots on Earth: Observations and Simulations in 2016

Johanna Yepes Universidad Nacional de Colombia, Sede Medellín, Facultad de Minas, Departamento de Geociencias y Medio Ambiente, Medellín, Colombia, and Department of Atmospheric Sciences, Desert Research Institute, Reno, Nevada

Search for other papers by Johanna Yepes in
Current site
Google Scholar
PubMed
Close
,
John F. Mejía Department of Atmospheric Sciences, Desert Research Institute, Reno, Nevada

Search for other papers by John F. Mejía in
Current site
Google Scholar
PubMed
Close
,
Brian Mapes Rosenstiel School of Marine and Atmospheric Sciences, University of Miami, Miami, Florida

Search for other papers by Brian Mapes in
Current site
Google Scholar
PubMed
Close
, and
Germán Poveda Universidad Nacional de Colombia, Sede Medellín, Facultad de Minas, Departamento de Geociencias y Medio Ambiente, Medellín, Colombia

Search for other papers by Germán Poveda in
Current site
Google Scholar
PubMed
Close
Free access

ABSTRACT

The diurnal cycle of precipitation and thermodynamic profiles over western Colombia are examined in new GPM satellite rainfall products, first-ever research balloon launches during 2016 over both sea and land, and numerical simulations with the Weather Research and Forecasting (WRF) Model. This paper evaluates the Mapes et al. mechanism for midnight–early morning coastal convection that propagates offshore: reduction of inhibition in the crests of lower-tropospheric internal waves. Shipborne balloon launches confirm the evening development of such inhibition by a warm overhang in saturation moist static energy (SMSE) near 700–800 hPa. This feature relaxes overnight, consistent with the disinhibition hypothesis for early morning rains. Over the coastal plain, soundings also show late afternoon increases in near-surface MSE large enough to predominate over the overhang’s inhibition effect, driving a second peak in the rainfall diurnal cycle. Parameterized convection simulations fail to simulate the observed coastal rainfall. Still, during a November 2016 wet spell, a cloud-permitting one-way nested 4 km simulation performs better, simulating morning coastal rainfall. In that simulation, however, early morning cooling in the 700–800 hPa layer appears mainly as a standing signal resembling the local radiative effect rather than as a propagating wave. We consider the additional hypothesis that the offshore propagation of that morning convection could involve advection or wind shear effects on organized convective systems. Strong easterlies at mountaintop level were indeed simulated, but that is one of the model’s strongest biases, so the mechanisms of the model’s partial success in simulating diurnal rainfall remain ambiguous.

Corresponding author: Johanna Yepes, ljyepes@unal.edu.co

ABSTRACT

The diurnal cycle of precipitation and thermodynamic profiles over western Colombia are examined in new GPM satellite rainfall products, first-ever research balloon launches during 2016 over both sea and land, and numerical simulations with the Weather Research and Forecasting (WRF) Model. This paper evaluates the Mapes et al. mechanism for midnight–early morning coastal convection that propagates offshore: reduction of inhibition in the crests of lower-tropospheric internal waves. Shipborne balloon launches confirm the evening development of such inhibition by a warm overhang in saturation moist static energy (SMSE) near 700–800 hPa. This feature relaxes overnight, consistent with the disinhibition hypothesis for early morning rains. Over the coastal plain, soundings also show late afternoon increases in near-surface MSE large enough to predominate over the overhang’s inhibition effect, driving a second peak in the rainfall diurnal cycle. Parameterized convection simulations fail to simulate the observed coastal rainfall. Still, during a November 2016 wet spell, a cloud-permitting one-way nested 4 km simulation performs better, simulating morning coastal rainfall. In that simulation, however, early morning cooling in the 700–800 hPa layer appears mainly as a standing signal resembling the local radiative effect rather than as a propagating wave. We consider the additional hypothesis that the offshore propagation of that morning convection could involve advection or wind shear effects on organized convective systems. Strong easterlies at mountaintop level were indeed simulated, but that is one of the model’s strongest biases, so the mechanisms of the model’s partial success in simulating diurnal rainfall remain ambiguous.

Corresponding author: Johanna Yepes, ljyepes@unal.edu.co

1. Introduction

Cloudy convection is a globally important process whose forced and free variations are at the heart of many weather and climate phenomena, in the tropics and beyond. Models struggle to replicate convection-dependent phenomena, especially models with cumulus parameterizations, motivating studies of convection’s mechanisms and behaviors in diverse geographical settings at various time scales. Diurnal variations can be especially telling, because convection's inherent process time scales are short enough to respond strongly to diurnal forcing, yet long enough that the resulting cycle of impactful variables like precipitation can be challenging to understand (Poveda et al. 2005; Ruppert 2016). Complex geography is also fruitful as a process probe, providing different natural laboratories in which the convective process is displayed and modulated (Biasutti et al. 2012). Observations are crucial links to reality in such complex settings, but there are never enough: models must also be brought to bear on interpreting the data. The catch has been that models struggle to replicate nature, as noted above—our very motivation, after all! Fortunately, with the growth of computing, explicit-convection models can increasingly succeed where parameterized-convection models fail, allowing fundamental mechanisms to be robustly diagnosed.

Many studies of coastal diurnal variations begin from the paradigm of land–sea breezes, adding moist convection as an embellishment. Radiative heating and cooling of Earth’s surface, imparted to near-surface air by direct contact, will induce circulations when gravity acts on the resulting density gradients. Although the Coriolis force (near-resonance with inertial motions) makes the problem counterintuitive near and beyond 30° latitude (Rotunno 1983), tropical sea breezes are direct and simple. It is tempting to think of land breezes as simply the opposite of a sea breeze, but there is significant asymmetry. Infrared cooling of the surface is not confined to nighttime, and is much weaker than daytime sunshine, especially under humid skies. Cooling is stabilizing rather than destabilizing, so the layer depths involved in day versus night can be very different. Stratified fluids respond to forcing in the form of internal or “gravity” waves, not simply as thermally direct overturnings. The implications of all this for coupling to moist convection are complicated enough to require quantitative modeling, guided by observations.

Careful thought and definition is also required: if nocturnal anomalies are defined as unoct = u(night) − [u(day) + u(night)]/2, then simple algebra dictates that half of unoct’s value is simply the negation of u(day). For instance, the “land breeze” to which night–morning convection off the coast of Borneo island was ascribed in the classic schematic of (Houze et al. 1981), reproduced as Fig. 1 of Mapes et al. (2003b, hereafter M03b) did not correspond to any actual offshore wind in the data, as M03b section 1b pointed out. That study went on to offer a counterhypothesis (discussed further below) for offshore nocturnal convection, based on simulations over western Colombia. From a mechanistic standpoint, the universal processes of convection motivating our study can be addressed by work even in different regions, and certainly in different years and seasons.

The far eastern Pacific and western Colombia feature one of the rainiest places in the world (Murphy 1939; Trojer 1958; López and Howell 1967; Arnett and Steadman 1970; Eslava 1993; Snow 1976; Meisner and Arkin 1987; Poveda and Mesa 2000; Sierra et al. 2015; Jaramillo et al. 2017; Yepes et al. 2019). Diverse oceanic and atmospheric processes converge there, from the planetary Hadley and Walker cells to a local cross-equatorial flow (southwesterlies) in the Choco low-level jet (ChocoJet) (Poveda and Mesa 1999, 2000; Poveda et al. 2006). The ChocoJet is characterized as a westerly flow that strengthens during September to November, whereas the northerly Panama jet intensifies during boreal winter (Xie et al. 2005). These two low-level jets are aspects of topographically channeled flows that surely include ascent in some places. Meanwhile, relatively drier easterlies cross the Andes at midlevels. Together with background vertical destabilization by near-equatorial sun and warm surface water (several degrees north of the equatorial cold tongue) (Raymond 2017), the region has many forcings and triggers and inhibitors that shape moist convection of various depths, scales, lifetimes, and timing (Raymond et al. 2003, 2006), including the strong activity of mesoscale convective systems (MCSs) (Velasco and Fritsch 1987; Mejía and Poveda 2005; Zuluaga and Houze 2015; Jaramillo et al. 2017). Few studies have focused on the diurnal cycle of precipitation over western Colombia. Satellite observations during the period of July–September 2000 (Mapes et al. 2003a) show an early morning peak over the coast that propagates offshore in the form of MCSs, although cell initiations appear to occur in a broad zone rather than at a single squall front. Larger remote sensing data surveys (Jaramillo et al. 2017) confirm this MCS development and propagation from land during afternoon hours to ocean during early morning hours, and farther offshore during the morning.

Based on MM5 model simulations, M03b proposed that the westward-propagating nocturnal offshore development of precipitating deep convection over the Colombian Pacific Ocean is released (i.e., disinhibited) by the cool phase (crest) of gravity waves forced by the diurnally oscillating heating of air over mountainous inland western Colombia. Similar waves were also shown to occur further south near the equator, suggesting that the coupling to deep moist convection is secondary rather than essential to their existence. The role of gravity waves in triggering offshore convection has also been studied in other tropical landscapes, with some debate about wave sources. For instance, Love et al. (2011) suggest that the late afternoon stratiform precipitation profile inland drives a gravity wave by midlevel evaporative cooling that destabilizes the offshore atmosphere later. Dry simulations over New Guinea described in Hassim et al. (2016) also indicate a significant role of inland convective systems, not just landscape sensible heating, in driving gravity waves that propagate over the ocean and destabilize (or disinhibit) conditions for convection. Using radiosonde observations, Yokoi et al. (2017) revealed that the low-level troposphere cools during afternoon just before the onset of precipitation in the western littoral of Sumatra, especially around intense convective events. Perhaps this too is caused by adiabatic ascent in a remotely forced internal gravity wave, although observations cannot discriminate mechanisms clearly.

This paper revisits M03b’s postulated mechanisms in western Colombia using observations and simulations from 2016. The observations were carried out during ChocoJEX (Yepes et al. 2019), the first field campaign to measure the ChocoJet’s dynamic and thermodynamic features. We first analyzed the thermodynamic profile observations, to establish a target for analyzing simulations. Then, we examined the diurnal cycle of precipitation in a series of three one-way (uncoupled) nested domains in the Weather Research and Forecasting (WRF) Model, comparing the results to new, better, campaign-coincident satellite data on precipitation.

The paper is organized as follows: section 2 presents the observational and model methods. Section 3 presents the field data (balloon soundings) by time of day, over land and sea and in different seasons. Section 4 evaluates the simulations against GPM satellite rainfall estimates and prior work, further interpreting the simulated processes with sensitivity experiments and evaluating the model against observations. Section 5 presents the summary and conclusions.

2. Data and model

a. ChocoJEX

This experiment (described in Yepes et al. 2019) measured dynamic and thermodynamic variables over the ChocoJet region. This low-level jet is a prominent feature of western Colombia climate that has been related to seasonal precipitation patterns (Poveda and Mesa 2000; Poveda et al. 2014; Bedoya-Soto et al. 2019). The experiment was carried out in 2016 and included two intensive observing periods (IOPs) over the ocean (in January and November) and two more inland (during June and October). It is important to mention that inland campaigns were carried out from a single venue at the city of Quibdó, whereas oceanic campaigns sampled mobile locations, following a vessel predesigned track. The number and location of soundings are summarized in Fig. 1 and Table 1. The collected variables included pressure, temperature, relative humidity, wind speed and wind direction using the MW41 Vaisala system with RS41-SGP radiosondes. Geopotential height was calculated hypsometrically from pressure and density profiles in the standard way.

Fig. 1.
Fig. 1.

ChocoJEX region of interest and IOP locations. Red (IOP1) and brown (IOP4) points over ship tracks represent individual sounding launching positions on board the route of the ARC Gorgona vessel. Inland soundings (IOPs 2 and 3) were launched from Quibdó (blue point). The inset corresponds to the three different domains used in this work. Modified from Yepes et al. (2019). The elevated terrain over Colombia and Ecuador correspond to the Andes Mountains.

Citation: Monthly Weather Review 148, 9; 10.1175/MWR-D-19-0405.1

Table 1.

Description of the four IOP developed during ChocoJEX (from Yepes et al. 2019).

Table 1.

b. Model setup and methods

The WRF Model (Skamarock et al. 2008), version 3.8.1, was used to perform a set of sensitivity experiments over the far eastern Pacific in the region delimited by 12°S–20°N and 95°–50°W, our domain 1 (d01, see Fig. 1). To assess resolution sensitivities, three one-way nested domains were simulated. The outer domain (d01) has 152 × 109 grid points with a 36 km grid size, the second domain (d02) has 237 × 195 grid points with a 12 km grid size, and the inner domain (d03) has 279 × 192 points with 4 km grid size. Since results are similar from d01 and d02, we contrast results from d02 and d03. All model domains have the same 49 sigma coordinate vertical levels with at least 25 levels distributed in the lower half of the troposphere. Initial and lateral boundary conditions were taken from European Centre for Medium-Range Weather Forecast interim reanalysis (ERA-Interim) data (Dee et al. 2011), which has 0.75° grid sizes and 6 h time increments. Of note is that the ChocoJEX soundings were not assimilated by any of the analysis and reanalysis systems. Hence, our observations constitute a source of independent data for model evaluation purposes.

The WRF Model physical parameterizations used the unified Noah land surface model (Chen and Dudhia 2001), Dudhia shortwave scheme (Dudhia 1989), Rapid Radiative Transfer Model (Mlawer et al. 1997) for longwave radiation, and the Mellor–Yamada–Janjić (Mellor and Yamada 1982; Janjić 1994) planetary boundary layer (PBL) scheme. The Kain–Fritsch (KF) cumulus parameterization scheme (Kain 2004) was used to parameterize convection for the two outer domains, whereas no convection parameterization was implemented for d03, hence convection and moist processes are treated explicitly. Two cloud microphysical (MP) schemes were compared: Thompson microphysics scheme (Thompson et al. 2008; hereafter referred as MP8), and the WRF single-moment 6-class scheme (Hong and Jim 2006; hereafter referred as MP6).

Each simulation was integrated between 1 and 30 November 2016. Model output was saved hourly, but the first 12 h of each simulation are regarded as spinup and not used in the analyses.

The diurnal cycle was constructed from hourly outputs by averaging all values at a particular UTC hour, as our analysis was focused on the Colombian Pacific region, all contained within one time zone. The performance of the simulated precipitation spatiotemporal distribution was assessed with GPM data (Huffman et al. 2018). We use the GPM IMERG products, which intercalibrate, merge and interpolate all satellite microwave precipitation estimates, together with microwave-calibrated infrared (IR) satellite estimates, into final values calibrated against precipitation gauges. The merged data product consists of half-hourly averages (summed in pairs for our hourly comparisons) and has a spatial resolution of 0.1° × 0.1°. Modeled rainfall was regridded to this same spatial resolution for all comparisons to GPM.

Longitude–time graphics of the GPM data in November 2016 (not shown) led us to select three contrasting periods of interest: dry days (15–21 November), no-propagation days (24–30 November), and propagation days (6–12 November). The propagation days were those with offshore migration of significant precipitation (up to 5 mm h−1 rates). The no propagation days exhibited convection confined over the coast, while the dry days had offshore and coastal precipitation less than 1 mm h−1.

c. The moist static energy (MSE) diagram

Moist static energy (MSE) plots were used to describe the diurnal thermodynamic variability during ChocoJEX campaigns. For each IOP, the four launch times (1900, 0100, 0700, and 1300 LT) temperature T and specific humidity q have been averaged over all available days (about 7 days per IOP). Diagrams of dry, moist, and saturated-moist static energy indicate the effects of environmental thermal perturbations on the buoyancy of a lifted parcel, which conserves MSE (h = CpT + gz + Lq) (Yanai et al. 1973) and thus follows a vertical line on the plot. If such a lifted parcel is saturated, its buoyancy is indicated by its position relative to the saturated MSE (SMSE or hsat) curve of the environment.

3. Observed diurnal features

In this section we present the diurnal cycles of precipitation as area-averaged composites for every IOP and their associated thermodynamic profiles. Further details concerning large-scale conditions during each IOP are presented in Yepes et al. (2019).

a. IOP1: 15–22 January 2016

The diurnal cycle of precipitation during this IOP is characterized by a single peak between 0700 and 0800 LT with values around 1 mm h−1 (Fig. 2). The convection during this IOP was scarce (satellite imagery at https://go.nasa.gov/2VrtYku) owing to El Niño conditions and the southernmost seasonal location of the ITCZ (Yepes et al. 2019).

Fig. 2.
Fig. 2.

Diurnal cycle of precipitation using GPM as an area-averaged over 3°–7°N, 77°–80°W during IOP1, over 2°–5°N, 78°–84°W during IOP4, and over Quibdó location (5.69°N, 76.65°) during IOP2 and IOP3.

Citation: Monthly Weather Review 148, 9; 10.1175/MWR-D-19-0405.1

In Fig. 3, an SMSE inversion layer (overhang of the SMSE curve) creates a layer where the lifted MSE-conserving parcel (vertical green line) has negative buoyancy at 1900 and 0100 LT in the 700–800 hPa layer. Although higher values of CAPE are observed at those hours, because surface air has higher MSE and thus more buoyancy at upper levels if lifted in an undilute manner, the negative buoyancy layer could inhibit surface-based convective clouds from deepening beyond the 700 hPa level. By 0700 LT (and at 1300 LT) the inversion layer weakens, allowing positive lifted parcel buoyancy at all levels. This could permit the deepening (vertical development) of convection, perhaps helping explain the morning rainfall peak observed over this area.

Fig. 3.
Fig. 3.

Static energy diagrams (in units of kJ kg−1) averaged by local hour (see panel titles) of the soundings during IOP1, including vertical profiles of dry static energy (black curve), moist static energy (MSE or h, blue curve), and saturation MSE (SMSE or hsat) (red curve). The family of gray curves indicates RH from 10% to 90%, and the blue shaded area is proportional to column integrated water vapor mass. The green curves show the MSE for lifted air parcels from the surface, for entrainment rates of 0 (heavy), 0.1, 0.2, and 0.5 km−1. Plotting code is from a free Python package called MSEplots, which in turn rests on Unidata’s MetPy package (https://github.com/weiming9115/MSEplots).

Citation: Monthly Weather Review 148, 9; 10.1175/MWR-D-19-0405.1

For this oceanic IOP, cooling at 0700 LT between 700 and 800 hPa is consistent with M03b results suggesting that adiabatic cooling by gravity wave signals may destabilize (or disinhibit) the atmosphere to morning convective development near the coast as it is shown in Fig. 2.

b. IOP2: 25 June–1 July 2016

During the first inland soundings (IOP2), some convective systems developed (satellite imagery at https://go.nasa.gov/2VOET6D) and the diurnal cycle showed a midnight peak with values below 1 mm h−1 (magenta in Fig. 2).

Soundings for this IOP again show an SMSE overhang between 700 and 800 hPa in the afternoon and evening (1300 and 1900 LT), which disappears at 0100 LT (Fig. 4). This earlier disappearance at the inland location of IOP2 (compared to offshore IOP1), and the earlier nighttime rain peak (Fig. 2), would be consistent with the notion of a westward propagating signal. Despite the prominent warm bulge and SMSE overhang at 700–800 hPa, late-afternoon (1700 LT) lifted parcels are nonetheless positively buoyant there because surface air MSE is so great at that hour. Relative humidity (RH) conditions (RH > 50% at all levels, as indicated by the blue fill area being wider than the white sliver between the blue and red curves) are suggestive of the impacts of frequent deep convection in this area, seen also in satellite images (https://go.nasa.gov/2VOET6D).

Fig. 4.
Fig. 4.

As in Fig. 3, but for IOP2 (see Fig. 1 for sounding location).

Citation: Monthly Weather Review 148, 9; 10.1175/MWR-D-19-0405.1

c. IOP3: 15–21 October 2016

The second inland campaign (IOP3) featured the most precipitation, with a two-peaked diurnal cycle (Fig. 2) at 0700 and 2000 LT, with values around 3 and 2 mm h−1, respectively. Thermodynamic profiles (Fig. 5) again feature the SMSE overhang in a diurnally varying warm bulge in the 700–800 hPa layer but its negative effect on lifted parcel buoyancy is compensated by the high-MSE (warm and moist) boundary layer conditions in the late afternoon. High RH (narrowness of white sliver) and satellite images (https://go.nasa.gov/2zpJbdr) evidence the frequent development of deep convection during the IOP.

Fig. 5.
Fig. 5.

As in Fig. 3, but for IOP3 (see Fig. 1 for sounding location).

Citation: Monthly Weather Review 148, 9; 10.1175/MWR-D-19-0405.1

d. IOP4: 21–28 November 2016

This was another offshore IOP, with deep convection suppressed especially in its early days by Hurricane Otto north of Panama (Yepes et al. 2019, satellite imagery at https://go.nasa.gov/2XW10Lb). Precipitation was near zero, with only a tiny 0100 LT maximum (Fig. 2). On average, low MSE air near the surface could not achieve lifted buoyancy at any altitude above 850 hPa (Fig. 6). In addition, SMSE overhangs near 800 and 500 hPa would limit vertical development of buoyant cumuli rooted in surface parcels with MSE exceeding the multiday mean. Above each of these overhangs, RH (the ratio of blue fill width to the white area between blue and red curves) drops sharply, consistent with this depth limiting effect on upward moisture transport by convection. RH drops to almost zero above the 500 hPa altitude. Diurnal modulation of these two SMSE warm layers is harder to see since they never disappear entirely at any hour, but careful overlay shows that 0700 LT is coolest in the lower troposphere as in the other marine IOP1 (Fig. 3).

Fig. 6.
Fig. 6.

As in Fig. 3, but for IOP4 (see Fig. 1 for sounding location).

Citation: Monthly Weather Review 148, 9; 10.1175/MWR-D-19-0405.1

Despite observational uncertainty related to undersampling and the moving platform during oceanic campaigns, the seasonal differences between IOPs, and synoptic variability, our observations confirm early morning cooling in the lower free troposphere that is consistent with the gravity wave mechanism, and apparently often strong enough to inhibit and disinhibit the buoyancy of updrafts in the 700–800 hPa layer. Robustness is indicated by the presence of this signal in averages that exclude even or odd days (not shown).

IOP4 was performed at the southernmost end of the Colombian Pacific, where drier conditions predominate climatologically as well as during this specific period. However, the ChocoJet was prominent in wind profiles during this IOP (Yepes et al. 2019), so modeling was focused on November. The next section will examine simulated mechanisms relevant to modulation of deep convection, focusing north of the ship location where more deep convection occurred.

4. Model results during IOP4

We performed a set of month-long simulations and model sensitivity experiments for a period encompassing IOP4. Synthetic soundings from the model are first compared to those observed. We then examine the model output more broadly, to compare and contrast our results with M03b’s simulations for a different model, season, and year.

a. Temperature, specific humidity, and winds

WRF Model output sampled at the sounding locations were contrasted against IOP4 upper-air observations in terms of mean temperature, specific humidity, and zonal and meridional wind bias profiles (Fig. 7). Both d02 and d03 exhibit similar biases (blue versus red curves), with a much larger sensitivity to the microphysics scheme (solid versus dotted curves) in d02 than in d03. Below 950 hPa a modest cool and dry bias (within the 95% confidence interval) prevails for MP8 microphysics; MP6 moisture bias is smaller. A moist bias near 800 hPa is outside the 95% confidence range and related to a cool bias in this same layer that might be characterized more pointedly as one lobe of a strongly unstable bias in lapse rate over the 900–800 hPa layer.

Fig. 7.
Fig. 7.

IOP4 biases of simulated profiles for (a) mean temperature, (b) specific humidity, (c) zonal, and (d) meridional wind. Simulated soundings were retrieved using the nearest grid point to each sounding location (see Fig. 1). Shaded gray areas correspond to difference intervals between model and observations with 95% of confidence using the Student’s t test.

Citation: Monthly Weather Review 148, 9; 10.1175/MWR-D-19-0405.1

Figure 7 shows a systematic northerly low-level wind bias around 900 hPa, and also a westerly bias (enhanced ChocoJet, Poveda and Mesa 2000) especially with MP6 microphysics. Significant easterly biases are evident aloft (750–400 hPa), likely due to smooth topography in the simulated Andes which relative to the real terrain would reduce the frictional forces and increase the wind. Upper-level meridional biases are relatively larger than low-level, but so was observational variability (increases significance range).

Diurnally, biases behave similarly, except a colder bias at 700 hPa during morning hours and stronger low-level westerlies during late afternoon (not shown). Surprisingly, there are larger differences between microphysics schemes than parameterized versus explicit convection (and grid size). However, explicit d03 was small and downscaled from d02, so many of its biases were arguably inherited directly from d02’s biases.

b. Precipitation

Figure 8a shows GPM precipitation estimates for November 2016. Two regions of intense precipitation occurred near the Colombian Pacific coast: an inland peak over 5°N, 77°W, and a much broader and more intense offshore band of rainfall spanning 3°–8°N. Figure 8b shows GPM-based November 2016 precipitation anomalies (relative to 2000–16), with anomalously low rainfall west of Colombia and especially high rainfall over most Northern Hemisphere lands and the Caribbean, as well as narrowly along the Pacific coasts of Panama and Costa Rica. Although the oceanic Niño index (ONI) was negative during October–November–December 2016, Fig. 8b is opposite to its typical effect of more rain over the Colombian Pacific coast (Poveda et al. 2011). Not all ENSO events and their impacts are equal (Trenberth and Smith 2006; Kao and Yu 2009), and we believe that November 2016s dry anomaly over the far EPAC was partly related to the presence of Hurricane Otto over the western Caribbean shifting monthly rainfall northward and closer to Central America.

Fig. 8.
Fig. 8.

November 2016 precipitation (mm day−1) using: (a) GPM mean, (b) GPM anomalies (relative to November, 2000–16), (c) simulated mean for KF + MP8 + d02, (d) bias for KF + MP8 + d02, (e) simulated mean for KF + MP6 + d02, and (f) biasKF + MP6 + d02 bias estimated relative to GPM. The white circle in (a) indicates the inland peak at 5°N, 77°W. The green spots indicate the Otto’s track on 22 Nov (tropical storm), 23 Nov (tropical storm), 24 Nov (major hurricane), and 25 Nov (tropical storm).

Citation: Monthly Weather Review 148, 9; 10.1175/MWR-D-19-0405.1

Precipitation biases in KF + MP8 + d02 and KF + MP6 + d02 relative to GPM are shown in Figs. 8d and 8f. Overall, both runs show very similar spatial patterns over land and coastal regions, with bigger differences over the Caribbean. Wet biases are more distinct over the Colombian and Venezuelan Llanos, and offshore of the Pacific side of Costa Rica and Panama. Both d02 simulations (Figs. 8c,e) show rain belts tightly confined over the western Andes foothills, with two precipitation hotspots over Colombian Pacific area that agree with GPM mean, one, over the Pacific coastal plain and another one right offshore. These striking features are the focus of our remaining results and discussion.

Figure 9 shows a longitudinal cross section of monthly precipitation averaged between 5° and 7°N, with topography illustrated in gray for reference. GPM shows two local peaks over Colombian Pacific region: (i) a maximum of 10 mm day−1 over the western Andes foothills (around 76.5°W) and (ii) a maximum of 15 mm day−1 right offshore near 78°W. Although the parameterized convection in d02 simulates two maxima (blue curves), precipitation over the coast is too weak while the inland peak is too strong, as seen above in Figs. 8c and 8f. In contrast, the explicit convection simulation (d03) shows a much sharper and stronger coastal and offshore peak, especially with MP6 microphysics. Jaramillo et al. (2017) relates the observed coastal rainfall to long-lasting MCSs propagating westward. However, in this relatively short simulation (30 days), there were an insufficient number of MCS events to robustly conclude anything about the role of convection organization.

Fig. 9.
Fig. 9.

Cross section of monthly precipitation (mm day−1) during November 2016 for WRF simulations and GPM observations averaged between 5° and 7°N. Shaded gray area illustrates the topography in the transect. Black boxes correspond to (a) offshore (81°–79°W), (b) coastal (79°–77.5°W), and (c) inland (77.5°–76°W) regions for areal averages discussed in the text.

Citation: Monthly Weather Review 148, 9; 10.1175/MWR-D-19-0405.1

Another feature that stands out in Fig. 9 is that the model systematically underestimates precipitation inland. Hong and Dudhia (2012) have found similar results using models in the “gray zone” of resolution (like our 12 km d02) where it is uncertain if convective parameterization should be expected to outperform explicit but under-resolved simulations. Far offshore, the parameterized d02 simulations are similar to observed precipitation, while d03 undersimulates precipitation and exhibits a small artifact at its lateral boundary.

c. Diurnal cycle

Figure 10 shows the mean diurnal cycle of precipitation from WRF simulations and GPM observations during long-term November and for 2016 averaged over the boxes in Figs. 9a–c. In GPM observations, sharp early morning increases peaked at 0700–0800 LT, slightly earlier inland and later offshore. That morning peak ramped down by around noon inland, and ramped down quite sharply at around 1500–1700 LT near the coast and offshore, where secondary afternoon peaks occurred just before the rampdown. Climatological data (magenta) mostly agree with 2016, but with higher values. Inland, the morning peak is before sunrise climatologically, earlier than the 0700 LT peak in November 2016.

Fig. 10.
Fig. 10.

Diurnal cycle of precipitation (mm h−1) during November 2016 for GPM (black line), MP8-d02 (blue line), MP8-d03 (red line), MP6-d02 (blue dashed line), and MP6-d03 (red dashed line) and the long-term (2002–16) November precipitation for GPM over (a) offshore, (b) coast, and (c) inland. Gray shadow represents one standard deviation for the November 2016 hourly data. Averaging regions are indicated in Fig. 9.

Citation: Monthly Weather Review 148, 9; 10.1175/MWR-D-19-0405.1

Model precipitation is much lower in all simulations, but the night–morning peak in the simulated diurnal cycle is present, a little earlier than the GPM peak. Offshore (Fig. 10a) and over the coast (Fig. 10b), during early morning all configurations underestimate peak values by 1 mm h−1 or more, with 3-h delayed respect to GPM. Of note is the high variability of the morning peak during November 2016 with standard deviations up to 4 mm h−1. Over land (Fig. 10c), GPM shows a bimodal diurnal distribution with a morning maximum around 0700 LT and a secondary afternoon maximum around 1900 LT. However, GPM data has shown that this inland morning peak is highly influenced by coastal processes (not shown), decreasing the morning peak as the box moves to east. In fact, this afternoon peak, directly influenced by solar radiation, has been well documented by other WRF studies (Mooney et al. 2017; Huang and Gao 2017). The parameterized convection simulations show only an afternoon maximum but two hours delayed, whereas the explicit convection simulations produce both peaks but fail to capture their amplitude. Surprisingly, the explicit convection simulations do not produce significantly better results with respect to parameterized convection of the Colombian Pacific coast precipitation diurnal cycle.

d. Comparison with Mapes et al. (2003b)

In this section we contrast our simulations against those from M03b. However, we can only compare diurnal mechanism diagnostics since the studies used different models (WRF versus MM5) and different integration periods (November 2016 versus portions of August–September 1998). For the purpose of intercomparison we focus the analysis of the offshore precipitation and propagation characteristics by compositing results during the wet spell observed in 6–12 November, hereafter called propagation days. We also contrast with composites of no-propagation days in 24–30 November and dry days during 15–21 November.

M03b argued that the propagation of precipitation observed at diurnal scales is associated with the westward moving internal gravity wave emitted by the diurnally oscillating heat source of the elevated mixed layer over the Andes. These gravity waves were seen in the form of temperature perturbations at 800 hPa, and their cold phase would help to increase the buoyancy of lifted parcels offshore and favor convection at early morning hours. This mechanism is the focus of our analysis.

Figure 11 shows the observed diurnal propagation signal using GPI (Warner et al. 2003) averaged during August–September 1998 (Fig. 11a) and GPM during November 2016 (Fig. 11). Figure 11b also shows the GPM long-term (2002–16) November diurnal cycle. West of 77.5°W (the vertical line), offshore rainfall peaks in night and morning hours (above the horizontal lines denoting midnight). The onset of rain appears to propagate westward at a speed comparable to the 15 m s−1 dotted reference line, although complicated contours bending with distance offshore make speed estimation ambiguous. GPM appears to lag the GPI signal in Fig. 11a by about 3 h, and its values are much larger (color legends differ tenfold). In all panels, a narrow peak near 2000 LT appears over the western slope of the western Sierra, well separated from the night–morning offshore rain. Another distinct peak just after midnight appears in the Magdalena valley (near 74°W at these latitudes).

Fig. 11.
Fig. 11.

Hovmöllers of the diurnal cycle of precipitation (mm h−1) averaged over the 3°–7°N band for (a) GPI product used in Warner et al. (2003) during August–September 1998–99, (b) long-term (2002–16) GPM precipitation, and (c) GPM precipitation during November 2016. Note that color bar extends up to 0.5 mm h−1 in (a) and up to 4 mm h−1 in (b) and (c). The top-right contour illustrates the topography and the red line replicates the 15 m s−1 propagation line in (a).

Citation: Monthly Weather Review 148, 9; 10.1175/MWR-D-19-0405.1

Windowing narrowly across the coast, with GPM as a reference, Fig. 12 shows the WRF-simulated mean diurnal precipitation during the propagation days (6–12 November), no propagation day (24–30 November) and dry days (15–21 November). We have selected the WRF single-moment 6-class scheme (MP6) microphysics case; Thompson microphysics scheme (MP8) keeps the precipitation much closer to the coast. During propagation days (Figs. 12a–c), GPM shows a broad midday maximum straddling the coast from 79° to 76°W. Just offshore, the predawn rain of climatology was lacking in 2016, as seen in Figs. 10a, 10b, 11b, and 11c with only a small morning occurrence well offshore (gray shading just touching the dashed reference line in Fig. 12a). Over land, a premidnight peak occurs over the western Sierra foothills (76°–77°W), appearing to persist into the midday second peak at that longitude.

Fig. 12.
Fig. 12.

Hovmöllers of the diurnal cycle of precipitation (mm h−1) during propagation days (6–12 Nov), no-propagation days (24–30 Nov), and dry days (15–21 Nov) for (a),(d),(g) GPM precipitation; (b),(e),(h) simulated precipitation from Ex + MP6 + d03; and (c),(f),(i) temperature perturbations (K) at 700 hPa from Ex + MP6 + d03 simulation, respectively. Data were averaged between 5° and 7°N. Temperature perturbations are calculated at 700 hPa and are relative to mean daily temperature at each longitudinal belt. The dashed black line corresponds to a reference line of 15 m s−1 propagation speed. Note that the onset of precipitation in GPM occurs approximately 3 h later than in simulations.

Citation: Monthly Weather Review 148, 9; 10.1175/MWR-D-19-0405.1

In contrast, the simulation shows the onset of the offshore precipitation propagation over the coast just after midnight, and its development offshore follows the dashed curve well. No-propagation and dry days entirely lack any coherent sloped features in this time–longitude space, as expected.

Figures 12c, 12f, and 12i show the simulated 700 hPa diurnal temperature perturbations longitudinal bands averaged between 5° and 7°N for Ex + MP6 + d03. Cross sections of anomalous temperature (not shown) maximize at 700 hPa instead of 800 hPa as M03b selected. There is a large standing component, with the air warmest in evening and coolest near sunrise at all longitudes, qualitatively consistent with the local accumulation of direct radiative heating. Sloped features parallel to the dashed curve are not especially evident.

Longitudinal cross sections at 0300, 0600, and 0900 LT are shown in Fig. 13, averaged between 5° and 7°N during propagation, no propagation and dry days for Ex + MP8 + d03. Cooling at 700 hPa occurs progressively with time, but any hypothesized internal wave connection to the anomalously cool air over land (darker gray shades) is not especially clear.

Fig. 13.
Fig. 13.

Cross sections of simulated temperature perturbations (K) averaged in the 5°–7°N band during propagation days (6–12 Nov) for Ex + MP6 + d03 simulation at (a) 0300, (b) 0600, and (c) 0900 LT; during no propagation days (24–30 Nov) (d) 0300, (e) 0600, and (f) 0900 LT and during dry days (15–21 Nov) at (g) 0300, (h) 0600, and (i) 0900 LT. Perturbations are calculated relative to mean daily temperature at each longitudinal belt. The black contours correspond to total cloud > 0.05g kg−1. White shadow represents the topography.

Citation: Monthly Weather Review 148, 9; 10.1175/MWR-D-19-0405.1

Other mechanisms may also be important. For instance, Yokoi et al. (2019) show that in addition to internal wave mechanisms, low-level wind shear is a key process for the regeneration of new convective cells helping the morning offshore migration of convection from Sumatra. As a preliminary evaluation, Fig. 14 shows the diurnal cycle of the 1.5–5 km wind shear over the coast during the three composites. The lowest values of wind shear are observed during the dry days (red), while wind shear between 12 and 16 m s−1 was simulated during propagation and no propagation days. For the propagation days, the diurnal cycle of wind shear has two peaks at morning and afternoon hours, and a sharp minimum before dawn. Might these 15% variations in shear inhibit or enhance some aspect of organized convection important to the offshore propagation? Or is this 12–16 m s−1 mean value itself (easterlies at midlevels) sufficient to explain westward motion of storms of whatever origin in the model? It is worth recalling that these midlevel easterlies are a radical model bias (Fig. 7c), and it is possible they may be distorting the model’s internal wave field significantly, for instance through shear ducting effects. More robust study would be necessary to consider these questions more fully. Overall, the model appears to lack the wave signatures of M03b, leaving some mystery to why it succeeded in simulating coastal convection propagating offshore after its near-midnight appearance (Fig. 12b).

Fig. 14.
Fig. 14.

Wind shear (m s−1) averaged over 5°–7°N, 78°W from the Ex + MP6 + d03 simulation for the propagation days (6–12 Nov), no propagation days (24–30 Nov), and dry days (15–21 Nov). Wind shear was estimated as the zonal wind difference between 1.5 and 5 km.

Citation: Monthly Weather Review 148, 9; 10.1175/MWR-D-19-0405.1

5. Summary and conclusions

We analyzed the diurnal cycle of thermodynamic profiles in ChocoJEX in situ soundings, the first atmospheric soundings ever gathered over the Colombian Pacific oceanic and land regions. During four week-long campaigns, distributed over 2016, observed thermodynamic profiles support M03b’s proposed mechanism suggesting that internal waves near 700–800 hPa modulate convection development through the strengthening and removal of a SMSE inversion or overhang that caps the layer of buoyancy experienced by parcels rising from the surface. For the ocean campaigns, the inversion layer in the 700–800 hPa layer weakens (corresponding to passage of the cool, upward displacement crest of the gravity wave) during early morning (0700 LT). Such elevated cooling occurs earlier in the inland campaigns.

A second diurnal factor is the near-surface source of rising parcels. This includes the MSE of near-surface air, which is greatest at the end of a day of solar energy input and least before dawn, after hours of surface radiative cooling. That cooling also stabilizes PBL conditions, inhibiting or at least not initiating the rising movement of parcels that could become convective clouds. Is the interplay of these two factors–lower-tropospheric inhibition/disinhibition favoring night–morning hours and near-surface warmth and moisture favoring late afternoon—sufficient to understand the diverse (including bimodal) diurnal cycles of precipitation observed during the field campaigns, through lifted-parcel buoyancy arguments? Or might mechanisms specific to organized MCS storms (like wind shear) be necessary to consider?

Simulations of November 2016 weather with WRF had strengths and weaknesses, from easterly wind biases during IOP4 to parameterized convection not raining over the coast enough at any hour of the day, and instead focusing too much on topographic slopes. Further one-way nesting of an explicit simulation produced better coastal rainfall, including its night-to-morning propagation offshore, yet did not exhibit very clear propagating internal wave T anomalies in the 700–800 hPa layer. Midlevel easterly winds, perhaps boosted by strong model bias, approached the 10–15 m s−1 speed of westward motion of rainstorms, potentially creating critical layers or ducts to distort internal waves of comparable speed, and also potentially moving storms westward by simple advection. The shear of those winds could also engage mechanisms unique to organized convection, beyond the simple lifted-parcel buoyancy considerations around random convection. Our results suggest that MP6 microphysics performed better than MP8 in this case, more realistically simulating the offshore precipitation maximum propagation over the Colombian Pacific region.

New GPM satellite rainfall data confirm diurnal cycle features in the region of study, such as the early morning peak in the Cauca inter-Andean valley. November 2016 was not typical of other Novembers, with less rain overall, later (morning instead of predawn) development of rain near the coast, and slower and unclear westward propagation with gaps and discontinuities.

Using observations, we confirmed the existence of transient internal wave modulation of lifted-parcel buoyancy as described by M03b. However, the November simulation produced similar coastal-zone rainfall signatures and propagations without clear propagating internal wave signatures, suggesting that other mechanisms such as the role of shear-related dynamical forces in the development and propagation of convective systems (Yokoi et al. 2019) should also be considered. Some remaining questions include: what modulates the intensity and vertical extent of the diurnal internal waves driven by oscillating heating of the Andes and other land features? Are the inland convective systems direct, advectively transported precursors of the coastal and offshore morning convection? Are other mechanisms necessary to consider, given the known fact that MCS development and propagation are part of the rainfall phenomenology? There is still opportunity for more systematic campaigns, and even simply more integrations of explicit convection (cloud-permitting) models, to improve and generalize our understanding of the diurnal cycle of precipitation over one of the rainiest spots on Earth, as a step toward understanding the globally important process of tropical convection more generally.

Acknowledgments

We are grateful with scientists, students, collaborators, and local volunteers from Universidad Nacional de Colombia at Medellín, Universidad Tecnológica del Chocó at Quibdó, DIMAR, FAC, DRI who participated in the planning, execution, and data gathering during ChocoJEX. We also thank Wei-Ming Tsai for helping us with the MSE plots through his open access Python package called MSE-plots. The work of Johanna Yepes was funded by COLCIENCIAS Doctorate Fellowship Program. COLCIENCIAS and the DRI and its Division of Atmospheric Sciences (DAS) partially supported J. F. Mejia. The work of G. Poveda is funded by Universidad Nacional de Colombia at Medellín, Colombia. The work of B. Mapes was supported by NASA NEWS program Grant NNX15AD11G and by the National Science Foundation under Grant 1639722.

REFERENCES

  • Arnett, A. B., and C. R. Steadman, 1970: Low-level wind flow over Eastern Panama and Northwestern Colombia. ESSA Tech. Memo. ERLTM-ARL 26, Air Resources Laboratories, 74 pp.

    • Crossref
    • Export Citation
  • Bedoya-Soto, J. M., E. Aristizábal, A. M. Carmona, and G. Poveda, 2019: Seasonal shift of the diurnal cycle of rainfall over Medellin’s Valley, central Andes of Colombia (1998–2005). Front. Earth Sci., 7, 92, https://doi.org/10.3389/feart.2019.00092.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Biasutti, M., S. E. Yuter, C. D. Burleyson, and A. H. Sobel, 2012: Very high resolution rainfall patterns measured by TRMM precipitation radar: Seasonal and diurnal cycles. Climate Dyn., 39, 239258, https://doi.org/10.1007/s00382-011-1146-6.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chen, F., and J. Dudhia, 2001: Coupling an advanced land surface–hydrology model with the Penn State–NCAR MM5 modeling system. Part I: Model implementation and sensitivity. Mon. Wea. Rev., 129, 569585, https://doi.org/10.1175/1520-0493(2001)129<0569:CAALSH>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dee, D. P., and Coauthors, 2011: The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553597, https://doi.org/10.1002/qj.828.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dudhia, J., 1989: Numerical study of convection observed during the winter monsoon experiment using a mesoscale two-dimensional model. J. Atmos. Sci., 46, 30773107, https://doi.org/10.1175/1520-0469(1989)046<3077:NSOCOD>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Eslava, J., 1993: Some particular climatic features of the Colombian Pacific region. Atmósfera, 17, 4563.

  • 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, 161175, https://doi.org/10.5194/acp-16-161-2016.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hong, S.-Y., and J.-O. Jim, 2006: The WRF single-moment 6-class microphysics scheme (WSM6). J. Korean Meteor. Soc., 42, 129151.

  • Hong, S.-Y., and J. Dudhia, 2012: Next-generation numerical weather prediction: Bridging parameterization, explicit clouds, and large eddies. Bull. Amer. Meteor. Soc., 93, ES6ES9, https://doi.org/10.1175/2011BAMS3224.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Houze, R. A., S. G. Geotis, F. D. Marks, 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, 15951614, https://doi.org/10.1175/1520-0493(1981)109<1595:WMCITV>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Huang, D., and S. Gao, 2017: Impact of different cumulus convective parameterization schemes on the simulation of precipitation over China. Tellus, 69A, 1406264, https://doi.org/10.1080/16000870.2017.1406264.

    • Search Google Scholar
    • Export Citation
  • Huffman, G., and Coauthors, 2018: NASA Global Precipitation Measurement (GPM) Integrated Multi-satellitE Retrievals for GPM (IMERG). Global Precipitation Measurement (GPM) National Aeronautics and Space Administration (NASA), accessed 25 July 2018, https://pmm.nasa.gov/sites/default/files/imce/times_allsat.jpg.

  • Janjić, Z. I., 1994: The step-mountain eta coordinate model: Further developments of the convection, viscous sublayer, and turbulence closure schemes. Mon. Wea. Rev., 122, 927945, https://doi.org/10.1175/1520-0493(1994)122<0927:TSMECM>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Jaramillo, L., G. Poveda, and J. F. Mejía, 2017: Mesoscale convective systems and other precipitation features over the tropical Americas and surrounding seas as seen by TRMM. Int. J. Climatol., 37, 380397, https://doi.org/10.1002/joc.5009.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kain, J. S., 2004: The Kain–Fritsch convective parameterization: An update. J. Appl. Meteor., 43, 170181, https://doi.org/10.1175/1520-0450(2004)043<0170:TKCPAU>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kao, H.-Y., and J.-Y. Yu, 2009: Contrasting eastern Pacific and central Pacific types of ENSO. J. Climate, 22, 615632, https://doi.org/10.1175/2008JCLI2309.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • López, M. E., and W. E. Howell, 1967: Katabatic winds in the equatorial Andes. J. Atmos. Sci., 24, 2935, https://doi.org/10.1175/1520-0469(1967)024<0029:KWITEA>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 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, 934947, https://doi.org/10.1002/qj.809.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 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, 799812, https://doi.org/10.1175/1520-0493(2003)131<0799:DPORIN>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 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, 830844, https://doi.org/10.1175/1520-0493(2003)131<0830:DPORIN>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Meisner, B. N., and P. A. Arkin, 1987: Spatial and annual variations in the diurnal cycle of large-scale tropical convective cloudiness and precipitation. Mon. Wea. Rev., 115, 20092032, https://doi.org/10.1175/1520-0493(1987)115<2009:SAAVIT>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mejía, J. F., and G. Poveda, 2005: Ambientes atmosféricos de sistemas convectivos de mesoescala sobre Colombia durante 1998 según la TRMM y el Re-análisis NCEP/NCAR. Rev. Acad. Colomb. Cienc. Ex. Fis. Nat., 29, 495514.

    • Search Google Scholar
    • Export Citation
  • Mellor, G. L., and T. Yamada, 1982: Development of a turbulence closure model for geophysical fluid problems. Rev. Geophys., 20, 851875, https://doi.org/10.1029/RG020i004p00851.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mlawer, E. J., S. J. Taubman, P. D. Brown, M. J. Iacono, and S. A. Clough, 1997: Radiative transfer for inhomogeneous atmospheres: RRTM, a validated correlated-k model for the longwave. J. Geophys. Res., 102 16 66316 682, https://doi.org/10.1029/97JD00237.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mooney, P. A., C. Broderick, C. L. Bruyère, F. J. Mulligan, and A. F. Prein, 2017: Clustering of observed diurnal cycles of precipitation over the United States for evaluation of a WRF multiphysics regional climate ensemble. J. Climate, 30, 92679286, https://doi.org/10.1175/JCLI-D-16-0851.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Murphy, R. C., 1939: The littoral of Pacific Colombia and Ecuador. Geogr. Rev., 29, 133, https://doi.org/10.2307/210063.

  • Poveda, G., and O. J. Mesa, 1999: La corriente de chorro superficial del oeste (del Chocó) y otras dos corrientes de chorro en Colombia: Climatología y variabilidad durante las fases del ENSO. Rev. Acad. Colomb. Cienc. Ex. Fis. Nat., 89, 517528.

    • Search Google Scholar
    • Export Citation
  • Poveda, G., and O. J. Mesa, 2000: On the existence of Lloró (the rainiest locality on Earth): Enhanced ocean-land-atmosphere interaction by a low-level jet. Geophys. Res. Lett., 27, 16751678, https://doi.org/10.1029/1999GL006091.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Poveda, G., and Coauthors, 2005: Diurnal cycle of precipitation in the tropical Andes of Colombia. Mon. Wea. Rev., 133, 228240, https://doi.org/10.1175/MWR-2853.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Poveda, G., P. R. Waylen, and R. Pulwarty, 2006: Annual and inter-annual variability of the present climate in northern South America and southern Mesoamerica. Palaeogeogr. Palaeoclimatol. Palaeoecol., 234, 327, https://doi.org/10.1016/j.palaeo.2005.10.031.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Poveda, G., D. M. Álvarez, and Ó. A. Rueda, 2011: Hydro-climatic variability over the Andes of Colombia associated with ENSO: A review of climatic processes and their impact on one of the Earth’s most important biodiversity hotspots. Climate Dyn., 36, 22332249, https://doi.org/10.1007/s00382-010-0931-y.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Poveda, G., L. Jaramillo, and L. F. Vallejo, 2014: Seasonal precipitation patterns along pathways of South American low-level jets and aerial rivers. Water Resour. Res., 50, 98118, https://doi.org/10.1002/2013WR014087.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Raymond, D. J., 2017: Convection in the east Pacific intertropical convergence zone. Geophys. Res. Lett., 44, 562568, https://doi.org/10.1002/2016GL071554.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Raymond, D. J., G. B. Raga, C. S. Bretherton, J. Molinari, C. López-Carrillo, and Z. Fuchs, 2003: Convective forcing in the intertropical convergence zone of the eastern Pacific. J. Atmos. Sci., 60, 20642082, https://doi.org/10.1175/1520-0469(2003)060<2064:CFITIC>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Raymond, D. J., C. S. Bretherton, and J. Molinari, 2006: Dynamics of the intertropical convergence zone of the east Pacific. J. Atmos. Sci., 63, 582597, https://doi.org/10.1175/JAS3642.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rotunno, R., 1983: On the linear theory of the land and sea breeze. J. Atmos. Sci., 40, 19992009, https://doi.org/10.1175/1520-0469(1983)040<1999:OTLTOT>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ruppert, J. H., 2016: Diurnal timescale feedbacks in the tropical cumulus regime. J. Adv. Model. Earth Syst., 8, 14831500, https://doi.org/10.1002/2016MS000713.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sierra, J. P., P. A. Arias, and S. C. Vieira, 2015: Precipitation over northern South America and its seasonal variability as simulated by the CMIP5 models. Adv. Meteor., 2015, 634720, https://doi.org/10.1155/2015/634720.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Skamarock, C., and Coauthors, 2008: A description of the Advanced Research WRF version 3. NCAR Tech. Note NCAR/TN-475+STR, 113 pp., https://doi.org/10.5065/D68S4MVH.

    • Crossref
    • Export Citation
  • Snow, J. W., 1976: The climate of northern South America. Climates of Central and South America, W. Schwerdtfeg, Ed., Elsevier, 295404.

  • Thompson, G., P. R. Field, R. M. Rasmussen, and W. D. Hall, 2008: Explicit forecasts of winter precipitation using an improved bulk microphysics scheme. Part II: Implementation of a new snow parameterization. Mon. Wea. Rev., 136, 50955115, https://doi.org/10.1175/2008MWR2387.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Trenberth, K. E., and L. Smith, 2006: The vertical structure of temperature in the tropics: Different flavors of El Niño. J. Climate, 19, 49564973, https://doi.org/10.1175/JCLI3891.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Trojer, H., 1958: Meteorología y Climatología de la vertiente del Pacífico Colombiano. Rev. Acad. Colomb. Cienc. Ex. Fis. Nat., 10, 199219.

    • Search Google Scholar
    • Export Citation
  • Velasco, I., and J. M. Fritsch, 1987: Mesoscale convective complexes in the Americas. J. Geophys. Res., 92, 95919613, https://doi.org/10.1029/JD092iD08p09591.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Warner, T. T., B. E. Mapes, and M. Xu, 2003: Diurnal patterns of rainfall in northwestern South America. Part II: Model simulations. Mon. Wea. Rev., 131, 813829, https://doi.org/10.1175/1520-0493(2003)131<0813:DPORIN>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Xie, S.-P., H. Xu, W. S. Kessler, M. Nonaka, S.-P. Xie, H. Xu, W. S. Kessler, and M. Nonaka, 2005: Air–sea interaction over the eastern Pacific warm pool: Gap winds, thermocline dome, and atmospheric convection. J. Climate, 18, 520, https://doi.org/10.1175/JCLI-3249.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yanai, M., S. Esbensen, J.-H. Chu, M. Yanai, S. Esbensen, and J.-H. Chu, 1973: Determination of bulk properties of tropical cloud clusters from large-scale heat and moisture budgets. J. Atmos. Sci., 30, 611627, https://doi.org/10.1175/1520-0469(1973)030<0611:DOBPOT>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yepes, J., G. Poveda, J. F. Mejía, L. Moreno, and D. C. Rueda, 2019: CHOCO-JEX: A research experiment focused on the Chocó low-level jet over the far eastern Pacific and western Colombia. Bull. Amer. Meteor. Soc., 100, 779796, https://doi.org/10.1175/BAMS-D-18-0045.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 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, 37453761, https://doi.org/10.1175/MWR-D-16-0468.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yokoi, S., S. Mori, F. Syamsudin, U. Haryoko, and B. Geng, 2019: Environmental conditions for nighttime offshore migration of precipitation area as revealed by in situ observation off Sumatra Island. Mon. Wea. Rev., 147, 33913407, https://doi.org/10.1175/MWR-D-18-0412.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zuluaga, M. D., and R. A. Houze, 2015: Extreme convection of the near-equatorial Americas, Africa, and adjoining oceans as seen by TRMM. Mon. Wea. Rev., 143, 298316, https://doi.org/10.1175/MWR-D-14-00109.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
Save
  • Arnett, A. B., and C. R. Steadman, 1970: Low-level wind flow over Eastern Panama and Northwestern Colombia. ESSA Tech. Memo. ERLTM-ARL 26, Air Resources Laboratories, 74 pp.

    • Crossref
    • Export Citation
  • Bedoya-Soto, J. M., E. Aristizábal, A. M. Carmona, and G. Poveda, 2019: Seasonal shift of the diurnal cycle of rainfall over Medellin’s Valley, central Andes of Colombia (1998–2005). Front. Earth Sci., 7, 92, https://doi.org/10.3389/feart.2019.00092.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Biasutti, M., S. E. Yuter, C. D. Burleyson, and A. H. Sobel, 2012: Very high resolution rainfall patterns measured by TRMM precipitation radar: Seasonal and diurnal cycles. Climate Dyn., 39, 239258, https://doi.org/10.1007/s00382-011-1146-6.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chen, F., and J. Dudhia, 2001: Coupling an advanced land surface–hydrology model with the Penn State–NCAR MM5 modeling system. Part I: Model implementation and sensitivity. Mon. Wea. Rev., 129, 569585, https://doi.org/10.1175/1520-0493(2001)129<0569:CAALSH>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dee, D. P., and Coauthors, 2011: The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553597, https://doi.org/10.1002/qj.828.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dudhia, J., 1989: Numerical study of convection observed during the winter monsoon experiment using a mesoscale two-dimensional model. J. Atmos. Sci., 46, 30773107, https://doi.org/10.1175/1520-0469(1989)046<3077:NSOCOD>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Eslava, J., 1993: Some particular climatic features of the Colombian Pacific region. Atmósfera, 17, 4563.

  • 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, 161175, https://doi.org/10.5194/acp-16-161-2016.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hong, S.-Y., and J.-O. Jim, 2006: The WRF single-moment 6-class microphysics scheme (WSM6). J. Korean Meteor. Soc., 42, 129151.

  • Hong, S.-Y., and J. Dudhia, 2012: Next-generation numerical weather prediction: Bridging parameterization, explicit clouds, and large eddies. Bull. Amer. Meteor. Soc., 93, ES6ES9, https://doi.org/10.1175/2011BAMS3224.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Houze, R. A., S. G. Geotis, F. D. Marks, 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, 15951614, https://doi.org/10.1175/1520-0493(1981)109<1595:WMCITV>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Huang, D., and S. Gao, 2017: Impact of different cumulus convective parameterization schemes on the simulation of precipitation over China. Tellus, 69A, 1406264, https://doi.org/10.1080/16000870.2017.1406264.

    • Search Google Scholar
    • Export Citation
  • Huffman, G., and Coauthors, 2018: NASA Global Precipitation Measurement (GPM) Integrated Multi-satellitE Retrievals for GPM (IMERG). Global Precipitation Measurement (GPM) National Aeronautics and Space Administration (NASA), accessed 25 July 2018, https://pmm.nasa.gov/sites/default/files/imce/times_allsat.jpg.

  • Janjić, Z. I., 1994: The step-mountain eta coordinate model: Further developments of the convection, viscous sublayer, and turbulence closure schemes. Mon. Wea. Rev., 122, 927945, https://doi.org/10.1175/1520-0493(1994)122<0927:TSMECM>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Jaramillo, L., G. Poveda, and J. F. Mejía, 2017: Mesoscale convective systems and other precipitation features over the tropical Americas and surrounding seas as seen by TRMM. Int. J. Climatol., 37, 380397, https://doi.org/10.1002/joc.5009.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kain, J. S., 2004: The Kain–Fritsch convective parameterization: An update. J. Appl. Meteor., 43, 170181, https://doi.org/10.1175/1520-0450(2004)043<0170:TKCPAU>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kao, H.-Y., and J.-Y. Yu, 2009: Contrasting eastern Pacific and central Pacific types of ENSO. J. Climate, 22, 615632, https://doi.org/10.1175/2008JCLI2309.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • López, M. E., and W. E. Howell, 1967: Katabatic winds in the equatorial Andes. J. Atmos. Sci., 24, 2935, https://doi.org/10.1175/1520-0469(1967)024<0029:KWITEA>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 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, 934947, https://doi.org/10.1002/qj.809.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 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, 799812, https://doi.org/10.1175/1520-0493(2003)131<0799:DPORIN>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 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, 830844, https://doi.org/10.1175/1520-0493(2003)131<0830:DPORIN>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Meisner, B. N., and P. A. Arkin, 1987: Spatial and annual variations in the diurnal cycle of large-scale tropical convective cloudiness and precipitation. Mon. Wea. Rev., 115, 20092032, https://doi.org/10.1175/1520-0493(1987)115<2009:SAAVIT>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mejía, J. F., and G. Poveda, 2005: Ambientes atmosféricos de sistemas convectivos de mesoescala sobre Colombia durante 1998 según la TRMM y el Re-análisis NCEP/NCAR. Rev. Acad. Colomb. Cienc. Ex. Fis. Nat., 29, 495514.

    • Search Google Scholar
    • Export Citation
  • Mellor, G. L., and T. Yamada, 1982: Development of a turbulence closure model for geophysical fluid problems. Rev. Geophys., 20, 851875, https://doi.org/10.1029/RG020i004p00851.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mlawer, E. J., S. J. Taubman, P. D. Brown, M. J. Iacono, and S. A. Clough, 1997: Radiative transfer for inhomogeneous atmospheres: RRTM, a validated correlated-k model for the longwave. J. Geophys. Res., 102 16 66316 682, https://doi.org/10.1029/97JD00237.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mooney, P. A., C. Broderick, C. L. Bruyère, F. J. Mulligan, and A. F. Prein, 2017: Clustering of observed diurnal cycles of precipitation over the United States for evaluation of a WRF multiphysics regional climate ensemble. J. Climate, 30, 92679286, https://doi.org/10.1175/JCLI-D-16-0851.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Murphy, R. C., 1939: The littoral of Pacific Colombia and Ecuador. Geogr. Rev., 29, 133, https://doi.org/10.2307/210063.

  • Poveda, G., and O. J. Mesa, 1999: La corriente de chorro superficial del oeste (del Chocó) y otras dos corrientes de chorro en Colombia: Climatología y variabilidad durante las fases del ENSO. Rev. Acad. Colomb. Cienc. Ex. Fis. Nat., 89, 517528.

    • Search Google Scholar
    • Export Citation
  • Poveda, G., and O. J. Mesa, 2000: On the existence of Lloró (the rainiest locality on Earth): Enhanced ocean-land-atmosphere interaction by a low-level jet. Geophys. Res. Lett., 27, 16751678, https://doi.org/10.1029/1999GL006091.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Poveda, G., and Coauthors, 2005: Diurnal cycle of precipitation in the tropical Andes of Colombia. Mon. Wea. Rev., 133, 228240, https://doi.org/10.1175/MWR-2853.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Poveda, G., P. R. Waylen, and R. Pulwarty, 2006: Annual and inter-annual variability of the present climate in northern South America and southern Mesoamerica. Palaeogeogr. Palaeoclimatol. Palaeoecol., 234, 327, https://doi.org/10.1016/j.palaeo.2005.10.031.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Poveda, G., D. M. Álvarez, and Ó. A. Rueda, 2011: Hydro-climatic variability over the Andes of Colombia associated with ENSO: A review of climatic processes and their impact on one of the Earth’s most important biodiversity hotspots. Climate Dyn., 36, 22332249, https://doi.org/10.1007/s00382-010-0931-y.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Poveda, G., L. Jaramillo, and L. F. Vallejo, 2014: Seasonal precipitation patterns along pathways of South American low-level jets and aerial rivers. Water Resour. Res., 50, 98118, https://doi.org/10.1002/2013WR014087.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Raymond, D. J., 2017: Convection in the east Pacific intertropical convergence zone. Geophys. Res. Lett., 44, 562568, https://doi.org/10.1002/2016GL071554.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Raymond, D. J., G. B. Raga, C. S. Bretherton, J. Molinari, C. López-Carrillo, and Z. Fuchs, 2003: Convective forcing in the intertropical convergence zone of the eastern Pacific. J. Atmos. Sci., 60, 20642082, https://doi.org/10.1175/1520-0469(2003)060<2064:CFITIC>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Raymond, D. J., C. S. Bretherton, and J. Molinari, 2006: Dynamics of the intertropical convergence zone of the east Pacific. J. Atmos. Sci., 63, 582597, https://doi.org/10.1175/JAS3642.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rotunno, R., 1983: On the linear theory of the land and sea breeze. J. Atmos. Sci., 40, 19992009, https://doi.org/10.1175/1520-0469(1983)040<1999:OTLTOT>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ruppert, J. H., 2016: Diurnal timescale feedbacks in the tropical cumulus regime. J. Adv. Model. Earth Syst., 8, 14831500, https://doi.org/10.1002/2016MS000713.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sierra, J. P., P. A. Arias, and S. C. Vieira, 2015: Precipitation over northern South America and its seasonal variability as simulated by the CMIP5 models. Adv. Meteor., 2015, 634720, https://doi.org/10.1155/2015/634720.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Skamarock, C., and Coauthors, 2008: A description of the Advanced Research WRF version 3. NCAR Tech. Note NCAR/TN-475+STR, 113 pp., https://doi.org/10.5065/D68S4MVH.

    • Crossref
    • Export Citation
  • Snow, J. W., 1976: The climate of northern South America. Climates of Central and South America, W. Schwerdtfeg, Ed., Elsevier, 295404.

  • Thompson, G., P. R. Field, R. M. Rasmussen, and W. D. Hall, 2008: Explicit forecasts of winter precipitation using an improved bulk microphysics scheme. Part II: Implementation of a new snow parameterization. Mon. Wea. Rev., 136, 50955115, https://doi.org/10.1175/2008MWR2387.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Trenberth, K. E., and L. Smith, 2006: The vertical structure of temperature in the tropics: Different flavors of El Niño. J. Climate, 19, 49564973, https://doi.org/10.1175/JCLI3891.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Trojer, H., 1958: Meteorología y Climatología de la vertiente del Pacífico Colombiano. Rev. Acad. Colomb. Cienc. Ex. Fis. Nat., 10, 199219.

    • Search Google Scholar
    • Export Citation
  • Velasco, I., and J. M. Fritsch, 1987: Mesoscale convective complexes in the Americas. J. Geophys. Res., 92, 95919613, https://doi.org/10.1029/JD092iD08p09591.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Warner, T. T., B. E. Mapes, and M. Xu, 2003: Diurnal patterns of rainfall in northwestern South America. Part II: Model simulations. Mon. Wea. Rev., 131, 813829, https://doi.org/10.1175/1520-0493(2003)131<0813:DPORIN>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Xie, S.-P., H. Xu, W. S. Kessler, M. Nonaka, S.-P. Xie, H. Xu, W. S. Kessler, and M. Nonaka, 2005: Air–sea interaction over the eastern Pacific warm pool: Gap winds, thermocline dome, and atmospheric convection. J. Climate, 18, 520, https://doi.org/10.1175/JCLI-3249.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yanai, M., S. Esbensen, J.-H. Chu, M. Yanai, S. Esbensen, and J.-H. Chu, 1973: Determination of bulk properties of tropical cloud clusters from large-scale heat and moisture budgets. J. Atmos. Sci., 30, 611627, https://doi.org/10.1175/1520-0469(1973)030<0611:DOBPOT>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yepes, J., G. Poveda, J. F. Mejía, L. Moreno, and D. C. Rueda, 2019: CHOCO-JEX: A research experiment focused on the Chocó low-level jet over the far eastern Pacific and western Colombia. Bull. Amer. Meteor. Soc., 100, 779796, https://doi.org/10.1175/BAMS-D-18-0045.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 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, 37453761, https://doi.org/10.1175/MWR-D-16-0468.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yokoi, S., S. Mori, F. Syamsudin, U. Haryoko, and B. Geng, 2019: Environmental conditions for nighttime offshore migration of precipitation area as revealed by in situ observation off Sumatra Island. Mon. Wea. Rev., 147, 33913407, https://doi.org/10.1175/MWR-D-18-0412.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zuluaga, M. D., and R. A. Houze, 2015: Extreme convection of the near-equatorial Americas, Africa, and adjoining oceans as seen by TRMM. Mon. Wea. Rev., 143, 298316, https://doi.org/10.1175/MWR-D-14-00109.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Fig. 1.

    ChocoJEX region of interest and IOP locations. Red (IOP1) and brown (IOP4) points over ship tracks represent individual sounding launching positions on board the route of the ARC Gorgona vessel. Inland soundings (IOPs 2 and 3) were launched from Quibdó (blue point). The inset corresponds to the three different domains used in this work. Modified from Yepes et al. (2019). The elevated terrain over Colombia and Ecuador correspond to the Andes Mountains.

  • Fig. 2.

    Diurnal cycle of precipitation using GPM as an area-averaged over 3°–7°N, 77°–80°W during IOP1, over 2°–5°N, 78°–84°W during IOP4, and over Quibdó location (5.69°N, 76.65°) during IOP2 and IOP3.

  • Fig. 3.

    Static energy diagrams (in units of kJ kg−1) averaged by local hour (see panel titles) of the soundings during IOP1, including vertical profiles of dry static energy (black curve), moist static energy (MSE or h, blue curve), and saturation MSE (SMSE or hsat) (red curve). The family of gray curves indicates RH from 10% to 90%, and the blue shaded area is proportional to column integrated water vapor mass. The green curves show the MSE for lifted air parcels from the surface, for entrainment rates of 0 (heavy), 0.1, 0.2, and 0.5 km−1. Plotting code is from a free Python package called MSEplots, which in turn rests on Unidata’s MetPy package (https://github.com/weiming9115/MSEplots).

  • Fig. 4.

    As in Fig. 3, but for IOP2 (see Fig. 1 for sounding location).

  • Fig. 5.

    As in Fig. 3, but for IOP3 (see Fig. 1 for sounding location).

  • Fig. 6.

    As in Fig. 3, but for IOP4 (see Fig. 1 for sounding location).

  • Fig. 7.

    IOP4 biases of simulated profiles for (a) mean temperature, (b) specific humidity, (c) zonal, and (d) meridional wind. Simulated soundings were retrieved using the nearest grid point to each sounding location (see Fig. 1). Shaded gray areas correspond to difference intervals between model and observations with 95% of confidence using the Student’s t test.

  • Fig. 8.

    November 2016 precipitation (mm day−1) using: (a) GPM mean, (b) GPM anomalies (relative to November, 2000–16), (c) simulated mean for KF + MP8 + d02, (d) bias for KF + MP8 + d02, (e) simulated mean for KF + MP6 + d02, and (f) biasKF + MP6 + d02 bias estimated relative to GPM. The white circle in (a) indicates the inland peak at 5°N, 77°W. The green spots indicate the Otto’s track on 22 Nov (tropical storm), 23 Nov (tropical storm), 24 Nov (major hurricane), and 25 Nov (tropical storm).

  • Fig. 9.

    Cross section of monthly precipitation (mm day−1) during November 2016 for WRF simulations and GPM observations averaged between 5° and 7°N. Shaded gray area illustrates the topography in the transect. Black boxes correspond to (a) offshore (81°–79°W), (b) coastal (79°–77.5°W), and (c) inland (77.5°–76°W) regions for areal averages discussed in the text.

  • Fig. 10.

    Diurnal cycle of precipitation (mm h−1) during November 2016 for GPM (black line), MP8-d02 (blue line), MP8-d03 (red line), MP6-d02 (blue dashed line), and MP6-d03 (red dashed line) and the long-term (2002–16) November precipitation for GPM over (a) offshore, (b) coast, and (c) inland. Gray shadow represents one standard deviation for the November 2016 hourly data. Averaging regions are indicated in Fig. 9.

  • Fig. 11.

    Hovmöllers of the diurnal cycle of precipitation (mm h−1) averaged over the 3°–7°N band for (a) GPI product used in Warner et al. (2003) during August–September 1998–99, (b) long-term (2002–16) GPM precipitation, and (c) GPM precipitation during November 2016. Note that color bar extends up to 0.5 mm h−1 in (a) and up to 4 mm h−1 in (b) and (c). The top-right contour illustrates the topography and the red line replicates the 15 m s−1 propagation line in (a).