Revelle (TOGA) Radar Report
For October 2011
S. Rutledge, T. Lang, and A. Rowe
Operations summary:
This report covers the Revelle radar operation for DYNAMO Cruise 2. The TOGA radar operated 24/7 with no problems. The Cruise 2 data has arrived back at CSU and the dataset is fully intact (Cruise 1 as well). The radar has been very stable and reliable. During Cruise 1 radar uptime was 95%, and this improved to 99% during Cruise 2. The various calibrations done over the course of the month have been repeatable. At this point in time we estimate the TOGA radar is about 1 dB high.
Science summary:
Fig. 1 shows the areal-averaged rainfall within a 150 km radius centered on the R/V Revelle. Lesser rain amounts were observed between 1-15 October, with heavier rain amounts in the second half of the month. There was a marked difference in the type of convection observed between 1-15 October and 16-30 October. (Note: The Revelle departed its station point on 29 October to make its port call in Phuket, Thailand). In general, isolated convective cells dominated the radar observations between 1-15 October. These cells generally drifted west to east. Most cells were shallow and dominated by warm rain production. Occasionally deeper convection was present, but remained isolated. In the latter half of the month, deeper convection was present, organized on the mesoscale. Large areas of stratiform precipitation were also observed.
Fig. 1. Areal-averaged rainfall observed by the Revelle radar between 1-30 October 2011. The DYNAMO Z-R was used. No attempt was made to discriminate between convective and stratiform rain. Analysis by Dr. Timothy Lang, CSU.
Examples of these isolated cells are shown in the following figures. Figure 2 shows a RHI from 1009 UTC on 11 October along 33 degrees azimuth. Echo tops were to 8 km, but these cells were largely warm rain producers given the low level echo centroids.
Fig. 2. RHI at 33 degrees azimuth from 1009 UTC, 11 October 2011.
An example of the scattered nature of cells around this period is shown in Fig. 3. During the first part of the month the sounding time series over the Revelle showed reduced upper tropospheric moisture and moderate tropospheric shear. Westerly winds persisted up to approximately 600 mb, explaining cell motion from west to east. Why mesoscale organization was generally absent in the first part of the month remains elusive, but the reduced upper level moisture and moderate shear are possibilities. (See Fig. 4).
We now turn to a brief discussion about the later half of the month. The next set of images is from the middle of the cruise when convective organization began to increase in an environment of directional vertical wind shear. Despite the increase in organization and overall radar echo depth, shallower, shorter-lived convection persisted. Fig. 5 shows these shallow and deeper modes of convection, as well as the sunrise. Fig. 6 shows the echo top product with impressive echo tops, marking onset of deeper convection around this time period. The corresponding RHI at that time (Fig. 7) highlights the deep embedded convection. Even in the shallower cells, reflectivities exceeded 50-55 dBZ.
During this period of increasing mesoscale organization, outflow boundaries played a key role in convective initiation, organization, and convective strengthening. One example occurred on 20-21 October, where colliding outflow boundaries resulted in the rapid intensification of convection near the radar (Fig. 8). On 24 October, an outflow boundary propagated over 100 km toward the east-southeast, later interacting with an ongoing MCS, leading to intensification of convection along a line embedded within the stratiform area (Fig. 9). In this case, the convective line continued to move toward the SE within the low-level westerly flow, while the stratiform area slowly extended back toward the NW, driven by the easterly winds above 600 mb on this day, although overall storm motion remained relatively slow. This trend of opposing directions of movement between the convective and stratiform components appeared to be a common characteristic of MCSs in this region forming within an environment characterized by weak winds, but directional vertical wind shear.
Another example is provided on 28 October, with enhancement of convection in an area where an outflow boundary converged with the prevailing low-level environmental flow (Fig. 10). On this day, wave after wave of squall type MCSs overran the Revelle moving northeast to southwest. Total rain at the ship on 28 October was 191.6 millimeters, the largest accumulation during cruise 2. Convection generally decayed after 29 October following the passage of the MJO and the onset of drying conditions.
Some examples of the extensive stratiform precipitation in the later portion of the month when extensive mesoscale systems were frequent are also
of interest. Fig. 11 shows a well defined bright band that seems to be extremely uniform. Note the sheared flow stratified by the melting level. Fig. 12 is another example, but this time the bright band is more convective in nature owing to the fact that this stratiform area formed via the decay of cells rather than the advection of ice particles over sloping trajectories from convection.
Fig. 13 shows a lightning Hovmoller plot using Vaisala's new GLD360 product, courtesy of Dr. R. Holle of Vaisala. The Hovmoller goes from the Equator to 10S to cover the DYNAMO area. First, the CG lightning patterns show the general system movements that have been previously documented. Convective elements moving west to east in the first part of the month, and then east to west after say 15 October. The first lightning event around 6-7 Oct was not really seen in any of the rain time series from Gan or Revelle. We are now analyzing daily maps to get a better handle on this (e.g., the lightning generating systems were south of the Gan/Revelle latitude). The zonally moving lightning signal around 14 Oct coincided with that peak rainfall on Gan Is. at that time. Revelle and Gan Is. both had a lot of rain during the 19-29 October period but the only significant lightning events were at the beginning and end of that period. Assuming that lightning indicates the presence of not just deep convection, but vertically developed convection (that is, cells are not only deep, 10 dBZ to > 16 km but the 30 dBZ contour extends well into the supercooled region) this might suggest that the deep and intense convection occurred near the start of the heavy rain period and was then again at the end. Monsoon-like rain falls between, where lightning is known to be infrequent at best. This is all very preliminary though. We are anxious to see if a similar pattern holds for the next MJO bursts.
In summary, a marvelous dataset was collected by the Revelle radar during cruise 2. We are anxious to combine the radar measurements with the ocean measurements made on the Revelle to begin addressing the air-sea coupling important to MJO initiation.
Fig. 13 Lightning Hovmoller. Courtesy Vailasa’s GLD360 dataset.