Michael Douglas *, Malaquias Peña, Walter Fernandez, Jose Luis Santos, Norma Ordinola and Luis Flores



The Pan American Climate Studies (PACS) Program of the National Oceanic and Atmospheric Administration (NOAA) has funded the establishment of a regional atmospheric SOunding NETwork (SONET) in Latin America to investigate the rainfall variability over Central America during the summer of 1997. The sounding network, PACS-SONET, consisted initially of 12 pilot balloon sites in Mexico, Nicaragua, Costa Rica, Panama, Colombia, Ecuador and Peru, together with one radiosonde station on Cocos Island in the eastern Pacific Ocean. These sites were established during April and May 1997 with the plan to operate for six months. Nearly all sites employed relatively inexpensive optical theodolites and pilot balloons to obtain wind soundings. The suitability of this relatively low-technology approach for the region was based on previous field program experience in Mexico. The observing program anticipated twice-daily observations, so that the amplitude of diurnal wind variations could be estimated. This would permit an assessment of the uncertainties inherent in climatological upper-air wind analyses that are generally based upon only once-daily soundings.

Shortly after the first stations were established in April 1997, it became apparent that a significant warm event in the eastern and central Pacific was developing. By mid-summer, the extreme magnitude of the event had become apparent. This led to two conclusions: 1) the observations being made during the summer of 1997 would be very anomalous over Central America, thus putting into question the extrapolation of potential scientific findings to other years, and 2) there existed the possibility of obtaining an unprecedented data set to describe the evolution of the wind field associated with the developing El Niño. Therefore, the observational plan was modified to: 1) establish additional pilot balloon stations in Ecuador and Peru to better sample conditions in the region of anticipated heavy rainfall, 2) extend the observational period at many stations through the end of October 1998, and 3) include the installation of several hundred inexpensive raingauges in coastal Ecuador and Peru to assure reasonable measurement of daily rainfall, for comparison with the wind soundings and with satellite observations.

Fig. 1 The PACS-SONET pilot balloon sites operating during the height of observations from Jan - June 1998.




In this section we highlight some of the findings from PACS-SONET that relate to the evolution of the monthly mean wind fields over the region, and how the year 1997 differed from 1998 and from long-term climatology.

2.1 Data

Most of the pilot balloon data were processed within each country and sent via Internet within a few hours to days of being collected. However, at some sites communications were less than ideal, and data were either not processed routinely, or were not sent frequently. Thus, months after the observations were made, the data sets for some sites are still not complete. Also, due to the different levels of training in how to detect and correct pilot balloon data errors, some bad data exists (typically at random levels in the vertical) in the data set that is now available from the PACS-SONET Homepage site (http://www.nssl.noaa. gov/projects/pacs). Despite these qualifications, monthly mean wind profiles have been calculated from the data that has been available, and these are presented here. The self-consistency of the monthly mean fields is obvious from the plots to be presented, such that the arrival of any late additional observations should not change the results.

The raw elevation and azimuth data were obtained every 30s up to 8 min after launch and thereafter every minute. From these values winds were calculated over 1-minute intervals below 8 minutes (about 2 km altitude) and over 2-min intervals above this level. Thus the wind profiles have somewhat higher vertical resolution in the lower troposphere than most radiosonde observations, whose winds are averaged over 2-min intervals.

Fig. 2. Monthly mean meridional wind profiles at San Cristóbal in the Galapagos Islands (1S) for period June 1997 to August 1998.

Although the number of observations diminished with altitude due to the blockage of clouds, good sampling was usually obtained up to 3 km level (~700 mb). This varies considerably with site and with season; in the discussion below we mention when the analyses are likely adversely affected by lack of data.

For comparisons with the observed monthly mean winds obtained from the pilot balloon observations we have used the National center for Atmospheric Research (NCAR)/National Centers for Environmental Prediction (NCEP) reanalysis products that are available from Internet sources (see Acknowledgments). These analyses (hereafter referred to as the NCEP reanalyses), produced using a fixed global data assimilation system, are available for an extended period and are being widely used for studies of the variations in atmospheric statistics on interannual time scales. We use these analyses here to compare with the observations that exist, and to point out areas where model output - observation differences are worthy of further investigation.

2.2 Mean meridional circulation near the equator

One of the most apparent aspects of the effect of the warm ocean temperature over the equatorial eastern Pacific associated with the 97-98 El Niño was the change in the near-surface winds at the equator. The monthly mean meridional winds at San Cristóbal (Fig. 2), based upon the pilot balloon observations made there, show a number of interesting aspects. Most evident during the period of June- November 1997 was the extreme steadiness of the lower-tropospheric flow. The maximum meridional wind, of about 7.5m/s from the south, occurred consistently at a height of about 300-400 m Above Ground Level (AGL) - also close to elevation above sea level. These southerly winds decrease rapidly with altitude, and by approximately 850 mb the meridional wind reverses. The meridional winds weakened during December 97, and during January, February, and March the meridional wind profile shows a very similar structure, with weak northerly flow and little vertical shear in the lowest several km. By April the v-wind has become weakly positive and by May the v-wind was nearly back to the values of the previous summer and fall.

The San Cristóbal observations show that the NCEP reanalysis products must be used with caution, even away from local topographic effects. Figure 3a shows a time-height section of the meridional wind at San Cristóbal based on the pilot balloon data; Fig. 3b shows the corresponding NCEP reanalysis section. Prior to comparing the sections two cautions are in order: 1) The sections extend up to only 800 mb because there are generally enough observations up to this level for reliable monthly mean values to be calculated and 2) the contouring in Fig 3b uses only data available at 1000, 925, 850 and 700 mb., even though the complete reanalysis dataset is available at finer vertical resolution. The most apparent difference between the two analyses is in the intensity of the meridional winds - during July-November 1997 the observations at 925 mb indicate a steady 4-5 m/s while the reanalysis indicates 2-2.5 m/s. Given that the lack of additional level data in the NCEP reanalysis prevents detecting the strong winds near 300 m AGL, it is still possible to compare the meridional wind changes between November 1997 and January 1998 at 925 mb. The difference in the observations is 6.5 - 7 m/s, while the reanalysis section shows only a 2 m/s decrease in meridional wind. Similarly, the increase in 925 mb meridional wind between March and May, 2.6 m/s, is less than the ~4.5 m/s indicated by observations. Comparisons with the full-resolution reanalysis data should be even more revealing.

Fig 3a. Time-height section showing meridional winds at San Cristóbal; winds from south are positive. Contour interval 1 m/s. v-component at 340 m (~970 mb) plotted to show maximum values. Vertical axis is pressure (mb).

Fig. 3b. Same as Fig. 3a except for NCAR/NCEP reanalysis for the same period. Contour interval .5 m/s.

2.3 Intensity of the trade winds over Central America

Time-height sections similar to that shown in Fig. 3 were prepared for the Managua mean pilot balloon observations and for an average of the 2 NCEP reanalysis grid points on either side of Managua to show fluctuations of the trade winds over Central America during the PACS-SONET (Fig. 4). Managua is located in a broad gap in the cordillera in Central America; the terrain is near sea-level and relatively flat. In addition, the cloudiness is less than at many other locations resulting in reliable mean wind from the pilot balloon data up to 3-4 km and often higher. Figure 4a shows that the strongest zonal flow (from observations) is near the 850 mb level, and during the months of July and August. The most apparent discrepancy between the observation-based section and the reanalysis section is that the winds in the latter are weaker (7 m/s for July and August 1997 vs ~ 14 m/s for observed) and the onset and decay of the strong winds during July and August is more abrupt. The two sections show similarities, such as the 850 mb maximum during December 1997, but the intensity is substantially underestimated in Fig. 3b and the additional observed maximum between February and March is not evident at all in the reanalysis section.

Fig 4a. Time-height section of zonal wind from monthly mean pilot balloon winds at Managua. Contour interval is 2 m/s. Values of selected maxima are plotted.

Fig. 4b. Same as Fig 4a except for NCEP reanalysis data. Note contour interval is 1 m/s.

Figure 5 highlights one particularly important feature of Fig 4. that has implications for regional climate variability studies. Inspection of Fig. 5 indicates that the zonal flow was about 2-3 m/s stronger during July-August 1997 than the same months of 1998, though the difference above 2.5 km altitude (about 750 mb) is less. However, Fig 4b suggests the opposite, or at best little difference in the zonal wind. If this is the magnitude of observed variability that can be expected between a dry year (1997) and a more normal year (1998) then reanalyses must be produced that reflect more closely the observations for them to provide useful insight. Naturally, comparisons with other sounding sites are desirable for a more definitive statement on the possible problems with the NCEP reanalyses in this region; the nearby sounding site at San Andres Island was unfortunately out of commission for most of 1997 and much of 1998.

2.4 Gap flow across the Isthmus of Tehuantepec

The pilot balloon station at Salina Cruz, on the south side of the Isthmus of Tehuantepec, provided the first above-surface measurements of the gap flow in this region. The most surprising finding has been the presence of a steady low-level northerly jet during the months of July and August (Fig. 6). Up to 1 km AGL the actual wind is from the north, thereafter turning to the east. This jet is not present in the NCEP reanalyses, which is not surprising as it is tied to the topography of the gap. The close similarity of the profiles during July and August suggests that the flow was steady during these months. Although there were considerable fluctuations in speed during July and August 1997, northerly jets were evident on a large percentage of the days during these two months.

Fig. 5. Zonal wind profiles at Managua for July-August 1997 and 1998.

A relatively rapid transition takes place between June and July (Fig. 7). June is characterized by weak vertical shear (at least of the meridional wind) and nearly calm conditions near the surface, whereas July shows the LLJ. Presumably the lower-tropospheric pressure difference increases substantially across the isthmus during this transition period.




More than 200 raingauge stations were established in northern Peru and western Ecuador, mostly with cooperative observers at Medical outposts or Civil Defense sites. These observations are being used to stratify the rainfall along the coast of northern Peru and Ecuador into wet and dry periods; this information, together with the pilot balloon data, will allow us to describe the synoptic flows associated with wet and dry spells. We hope to report on the special rainfall measurements made in Peru and Ecuador during the period of the heaviest El Niño rains, but at the moment the rainfall data is still being compiled from the widely spaced volunteer observers.

Fig. 6. The July-August 1997 monthly mean meridional wind profiles at Salina Cruz.

Fig. 7. June and July profiles of meridional wind at Salina Cruz.


The PACS-SONET activities were funded through the support of the NOAA PACS program. In addition, for their support in obtaining funds on short notice for the special El Niño observations, personnel from the PACS program (Andrea Ray, Mike Patterson) and numerous others from the NOAA Office of Global Programs are sincerely thanked. The support of many members of NSSL was essential for managing the numerous financial transactions, shipping activities, and paperwork required to complete the project. Rosario Douglas arranged countless details associated with the project both in the field and throughout the project. The support of almost innumerable observers, administrators and other meteorological service personnel in the various countries of the SONET domain is sincerely appreciated.

A more thorough list of acknowledgments (as well as other information and all of the soundings made during SONET) can be found on PACS-SONET Homepage address given in Sec 2.