THE ROLE OF SEA SURFACE TEMPERATURE AND VEGETATION CHARACTERISTICS IN THE SEASONAL EVOLUTION

OF SUMMER RAINS OVER NORTHWESTERN MEXICO

First-year’s Progress Report

December 2004

Principal Investigator: Michael W. Douglas Nacional Severe Storms Laboratory, Norman, Oklahoma.
CO-PI: Christopher Watts -Universidad de Sonora, Hermosillo, México.
CO-PI: Russel Scott - USDA-ARS, Tucson, Arizona

Progress report by: Jose Galvez - Cooperative Institute for Mesoscale Meteorological Studies (CIMMS) / University of Oklahoma

Abstract

Previous observations have shown that a seasonal march of the region of maximum precipitation from the slopes of the Sierra Madre Occidental (SMO) towards the Gulf of California (GOC) in Mexico occurs as the summer progresses. This project evolved to improve the understanding of the mechanisms involved in this migration. We speculate that this seasonal progression is caused by the changes in the vegetation characteristics during the onset period of the North American Monsoon (NAM) together with the seasonal increase in the GOC SSTs.

Diverse station networks were deployed before and during the first stages of the North American Monsoon Experiment (NAME) Field Campaign, carried out in the summer of 2004. The networks included one flux tower, 7 temporary pilot balloons stations, a network of simple raingauges, and a special sounding site at Tezopaco, Sonora where thermodynamic soundings and additional surface observations were made. The entire network operated during a 4-month period, from June through September, with the exception of the flux tower, which started operation on July 10th.

The special pilot balloon sounding network was designed to better describe circulation characteristics associated with the diurnal cycle, to describe the characteristics of the seasonal evolution of the sea-land breeze that occurs on the east side of the GOC, and to complement the other sounding sites installed in western and northern Mexico for the NAME. The pilot balloon soundings were made twice daily and synchronized with the existing upper air networks. The thermodynamic observations at Tezopaco were made together with the surface observations during four intensive observation periods (IOPs) designed to capture in more detail different stages of the NAM.

The flux tower, equipped with meteorological sensors at different levels, was installed at a dry forest site about 6 km east of Tezopaco, in southeast Sonora. It was designed to measure the evolution of the canopy-top fluxes associated with the vegetation changes of the tropical dry forest. The flux data have been collected almost continuously since the date of installation. The raingauge network consists of about 300 simple raingauges. These rain gauges are designed to measure daily rainfall about the central GOC region, thus they were deployed both in parts of the Baja California peninsula and in the coastal and foothill regions on the eastern side of the GOC. Both the tower sensors and the raingauge network continued their operations after 30 September 2004.

Except for the rain gauge network data, all of the different data sets were gathered in November 2004. By the time this report was created most of the data (except those from the flux tower) were still in a raw format given the short time since they were brought to the NSSL. Since the quality control process is just starting, the results currently available are preliminary. However, we are including some of these findings on this document, in particular the ones related to the thermodynamic soundings and the evolution of the dry forest foliage. This document ends with a statement of the next year’s planned activities.

1. Project activities carried out during 2004

1.1. The observation network

In order to explain the observed migration of precipitation towards the coast that occurs as the rainy season progresses, the changes of the vegetation, boundary layer structure, and diurnal circulations near the central Gulf of California (GOC) were quantified through observations. The array of stations operated during a four-month period, starting in late May 2004 and ending around 30 September 2004 (Table 1.1). Some of the stations continued operations after this date. Figure 1 shows a map of the entire network.

The special observational array installed and operated for the project purposes included a flux tower, an array of 7 pilot balloon sites, a raingauge network consisting of about 300 simple raingauges, and a radiosonde and tethered balloon sounding site. In addition to these observations, cloud photography, vegetation photography and surface observations were made in Tesopaco, Sonora during four Intensive Observation Periods (IOPs) carried out during the NAME. Each component of the network will be described in more detail below.

Table. 1.1. Field Campaign Calendar. The first column shows the dates organized in weekly intervals. The second column shows the pilot balloon network calendar. The third column shows the raingauge network calendar. The fourth displays the flux tower calendar and the fifth the IOP calendar.

A total of 2353 observations are available at the NSSL so far. Part of the data from Huatabampo as well as the raingauge data is not available yet. Table 1.2 summarizes this information.

Table. 1.2. Number of observations collected during the NAME Field Campaign for the project. The locations were the observations were taken are organized in column 1. observations have been stratified in 30-gram pilot balloon soundings, radiosonde soundings, tethersonde soundings and small balloon soundings. The totals per type, station and overall total are also included. The information for the site of Huatabampo is not complete yet.

Sounding Network

Six (6) pilot balloon stations were temporarily set up around the central region of the Gulf of California. Two sites were established on the eastern side of the Baja California Peninsula and four sites were set-up in Sonora and Sinaloa. The PACS-SONET (http://www.nssl.noaa.gov/projects/pacs/) operational pilot balloon station at Topolobampo modified the schedule of observations to match the one designed for the NAME temporary sites, adding to a total of 7 upper air stations utilized to fulfill the project goals.

The purposes of the sounding network were to better describe circulation characteristics associated with the diurnal cycle, describe the characteristics of the seasonal evolution of the sea-land breeze that occurs on the east side of the GOC, and finally to complement the observations of the Special Sounding Network installed in Mexico for the NAME.

The pilot balloon (“pibal”) measurements were made by following 30-gram balloons with optical theodolites (Figure 1.2). The assumed ascent rate of the balloons was previously estimated through double theodolite tests and found to be 3.61 m/s. The balloons were filled with helium. All of the equipment necessary for the measurements, including the forms, was supplied by the project. An amount of $10.00 USD was budgeted per observation. At most of the sites, except those operated by the Mexican Navy (Topolobampo), where there was no cost involved.

The establishment of the pilot balloon sites began in mid-May and was completed by early June. The observations were made twice daily, from the date of set-up of the site until September 30^th. The morning observation was made at 14 UTC and the afternoon observation at 23 UTC. The schedule was designed to describe the diurnal cycle of the local sea (or slope) breeze, taking into account the requirement that the observations be made during daylight (to simplify tracking the balloon with the theodolites) throughout the entire 4 months of observations.

Figure. 1.1. Station network configuration.

Three pibal sites were located on the coastal plain east of the GOC: Empalme, Huatabampo and the PACS-SONET site of Topolobampo. At Empalme, a Mexican National Weather Service radiosonde station, the observations were carried out in addition to the routine radiosonde observations. At Topolobampo, oceanographic staff of the Mexican Navy continued making observations twice a day, as they have during the previous four years. The Huatabampo site was installed at the Instituto Tecnologico de Huatabampo (ITHUA) and the observations were carried out twice a day except during Intensive Observation Periods (IOPs), when more frequent observations were made.

Two inland stations were established at Tezopaco and Choix. Both of these locations are at a relatively low altitude (~ 450 m ASL) and were considered as ideal to describe the intensity of the afternoon slope breezes towards the mountains. Both locations are also surrounded by dry forest that rapidly develops foliage after the first intense monsoon rainfall event. It was expected that these sites would show changes in the diurnal breezes associated with the changing vegetation characteristics.

Figure. 1.2. Preparing to launch a pilot balloon and track it with an optical theodolites at Tezopaco, Sonora.

Two coastal pilot balloon sites were set up in Baja California, at Loreto and Santa Rosalia. These sites were to help describe the sea-breeze evolution throughout the summer in a region less affected by summer rainfall – the region was more desertic than the other site of the GOC. Thus, the sea breeze changes are these two sites are expected to be mainly due to changes in SST – land temperature contrasts, not vegetation cover changes. These sites were also needed to calculate vertical motions (via continuity equation) over the Gulf of California and their seasonal evolution.

Four (4) IOPs were carried out during the NAME, during which the sites of Tesopaco and Huatabampo enhanced their schedule of observations by performing at least five (5) pilot balloon soundings per day. The length of the IOPs varied from a minimum of 5 days to a maximum of 14 days, with a “sub-IOP” that lasted two months.

Flux tower

A flux tower was set up for micrometeorological studies about 6 km east of Tesopaco, in the southern part of the state of Sonora (Figure 1.3). The main objective of the tower is to measure the energy and moisture fluxes before, during, and after the appearance of the foliage in the dry forest along the western foothills of the Sierra Madre Occidental (SOM).

Figure. 1.3. The flux tower.

Figure. 1.4. The bottom of the flux tower after rains have started.

The tower location is totally surrounded by forest. Since the canopy has an average height of about 8-10 meters, the tower utilized was 18 meters high to measure the fluxes below, inside, and above the canopy. The data collected include temperature, relative humidity, shortwave radiation, and fluxes of heat, momentum and moisture (SEE THE REPORT BY WATTS FOR DETAILS). The measurements started in early July 2004 and will be continued through a second rainy season in 2005. This tower is a potential candidate for continuous monitoring of the tropical dry forest environment. This part of the project is being coordinated by the Chris Watts from the Universidad de Sonora in Hermosillo, Mexico.

Raingauge Network

The network of simple raingauges was installed to describe the space and time evolution of rainfall by recording daily precipitation. The installation campaign started in late May, well before the beginning of the rains. About 250 raingauges were distributed among 9 municipalities in southern Sonora and northern Sinaloa. Additional raingauges were distributed in Baja California, close to the pilot balloon stations of Loreto and Santa Rosalia.

Due to time constraints, the raingauges were separated into groups and distributed among the municipalities, via meetings with the authorities of different local institutions. Sometimes the volunteer observers were also present. A short presentation and training course was provided at each meeting. Maps with ideal locations for the raingauges were presented and discussed. The final location of the gauges was determined with help and suggestions from the local authorities. At the end of the meetings the rain gauges were distributed among local volunteer observers, including a form designed to collect the data until October 31 2004. The form for recording the data was especially designed to minimize errors in reading the raingauges.

Figure. 1.5. Talks were presented in each municipality to help distribute the rain gauges.

Radiosonde sounding site at Tezopaco

A radiosonde observing site was established at Tezopaco, a small town located in southeast Sonora. Tezopaco is an inland location located just west of the foothills of the SMO, at an elevation of 440 m ASL. The dominant vegetation is tropical dry forest that rapidly develops foliage after the first intense rain events, so the location was found to be ideal for thermodynamic soundings, cloud, and vegetation photography. Tezopaco is about 6 km away from the flux tower location.

A pilot balloon site operated continuously from 20 May 2004 through 30 September 2004 making observations twice daily at 14:30 UTC and at 23 UTC. In addition to these wind profiles, tethersonde and radiosonde observations were made during four IOPs carried out at different stages of the warm season. Additional observations made during these periods included surface hourly observations and hourly cloud photography. In addition, 17 sites near Tezopaco were selected to document the evolution of the vegetation via photography. Some of the results are shown in this document.

Table 1.3. Schedule of observations for third IOP. The hours are in local time.

The calendar showing the times when the IOPs were carried out is shown in Table 1.1. The first IOP started on 3 June 2004 and ended on 11 June 2004 for a total of 9 days. It was designed to capture the prevailing atmospheric conditions before the onset of the monsoon. The second IOP lasted 5 days from June 29 to July 3, also before the rains. The third IOP lasted 14 days from July 14 to July 27, capturing a rainy period during the onset of the monsoon. The last IOP lasted almost 2 months from early August through September 30. The schedule of observations during this period was reduced compared to the prior IOPs mainly due to the length of the IOP.

The type and number of observations carried out during the first three IOPs was similar, with some minor modifications. Table 1.3 shows the schedule followed during IOP 3. During each IOP day about 13 atmospheric soundings were made, two of which were radiosonde soundings. The radiosondes were launched in the early morning and in the early afternoon to describe atmospheric thermodynamic profiles associated with the diurnal cycle. Tethersonde observations (Figure 1.6) were carried out during the mornings to measure the growth of the boundary layer before convective storm development. Cloud photography and surface observations of temperature and cloudiness were made on an hourly basis.

The vegetation changes during the summer were also documented through digital photography. For this purpose, 17 points located along a road extending southeast from Tezopaco were selected for monitoring the evolution of the foliage. Before the rains the pictures were taken at 3 to 4 day intervals. When the rains started (IOP3) the pictures were taken at two-day intervals since the changes in the foliage occurred rapidly. Some results are shown in figures 3.4 and 3.5.

In addition to Tezopaco, frequent observations were also made at Huatabampo during the IOPs. This site, located about 10km from the GOC on a very flat and extensive coastal plain, was used to describe the sea breeze and land breeze circulations over the coastal plain at the same time as the frequent Tezopaco soundings. A mix of 30 gm and smaller “party” balloons were used at this site and also at Tesopaco to save gas, which was often in short supply due to logistical problems.

1.2. Installation Campaign

The installation campaign started in 19 May 2004 and ended by early July 2004. The first stations set up were the pilot balloon sites in Sonora. The Sonoran raingauge network was established simultaneously by another team. The pilot balloon sites to be installed last were the ones in Baja California, which started their operations in early June. The last component to be installed was the flux tower, which started operations by 10 July 2004. The delay in setting up the flux tower was due to the late arrival of funds to do this (primarily buying the tower locally), in fact the funds were borrowed from another account while the OGP funds arrived. (The funds were actually “lost” within the University of Sonora, rather than arriving very late).

Before the establishment of the pilot balloon stations and the raingauge network a number of arrangements had to be made, such as arranging for balloon inflation gas and finding suitable observers at the desired sites. The set up of each pilot balloon site consisted of finding the ideal location within the area of interest, informing potential participants about the project through short talks and information sheets, finding potential observers interested on making the observations and providing at least three detailed training sessions. The training sessions were divided into short talks and long practices in the field. All of the equipment, which included optical theodolites and tripods, balloons, helium regulators and other accessories were provided. Two to three gas cylinders to start the observations were also provided per site with the exception of Empalme, where a hydrogen generator existed.

Fig. 1.6. Tethersonde and pilot balloon observations at Tesopaco, Sonora.

The establishment of the raingauge network started in late May. Each individual raingauge could not be installed by the team due to time constraints. Instead, the raingauges were grouped and distributed among 9 municipalities in Sonora, to be further installed by local individuals. Proteccion Civil from Sonora coordinated each and every meeting carried out at the municipalities. The installation team participated in each meeting, which consisted on an introduction to the project via talks and pamphlets, and further discussion with the authorities to decide the potential location of the raingauges and to find an institution responsible for their installation. A how-to-install and how-to-operate training session was also provided, including a manual, the raingauges, and the forms to collect the data.

The radiosonde and tethersonde site was established at Tezopaco and were operated by graduate students from the University of Oklahoma and by students from Xalapa, Mexico during the IOPs. Although minimal training was required at this site, the tethersonde winch bought with funds for this site did not arrive in time, since the funds arrived late at NSSL and the company building the winches was also late in delivery. Fortunately, another winch, older but still working well, was obtained on loan with the help of Rene Lobato at IMTA. Unfortunately, since none of the students who operated the winch had used one before there were many small complications that were discovered only through experience (The PI was unavailable elsewhere in Mexico to provide guidance on the use of the winch, except via phone, which wasn’t always adequate since Tezopaco did not have cell phone coverage). These led to some balloons being lost and other accidents occurring that complicated the data collection activity.

1.3. Collection of the equipment and data

The travel to collect all of the equipment deployed during the observational campaign started on 16 October 2004 and ended by 15 November 2004. All the materials used for the entire NAME Enhanced Upper Air Network, including the raw data, were collected and brought to the United States. The raingauge network data could not be gathered by 30 November 2004 due to the diversity of the observers, locations and especially accessibility. Proteccion Civil from Sonora is organizing the collection of these data.

2. Current status

Most of the information has already been collected, but requires organization and further quality control procedures, in particular the pilot balloon data. The rainfall data is not available yet. A description of the status of the different stations and networks follows.

Pilot Balloon Network and Data

By the time when this document was written, all the provisional pilot balloon sites had been removed, and the equipment used for the field campaign all brought back to the United States. The pilot balloon information had all been collected and brought to the NSSL in different formats. Even though some of the data were already in electronic format, most of them were still in sheets of paper. While some of this data is being run through quality control procedures, most of it still needs to be keypunched and put in an electronic format.

Flux tower

The flux tower is currently operating continuously. Data is being downloaded every week or two by personnel at ITSON in Obregon.

Raingauge Network and Data

Given the size of the network (~300 raingauges distributed in at least 10 different municipalities) and the diversity of the observers, locations and especially accessibility, the rainfall data could not be made accessible by 30 November 2004. At the time this document was created, Proteccion Civil from Sonora was in charge of the rainfall data collection process, with concentrated efforts on gathering not only the data but the exact location of each raingauge.

Tethersonde and Radiosonde Data

The tethersonde and radiosonde data has all been collected and is already in digital format. Even though some quality control programs have already been developed, they still need to be adjusted to some of the data sets and then applied.

Cloud and vegetation photography

The data, which is stored in RGB JPEG images, need only to be organized. The vegetation pictures need to be cropped into equal domains for comparison.

3. Preliminary Results

In this section of the report we show some preliminary results. In particular, we have chosen to show the evolution of the thermodynamic profiles made at Tezopaco and the changes in vegetation as documented by the digital photography. Figure 3.1, 3.2 and 3.3 show the thermodynamic profiles and therefore boundary layer evolution throughout the rainy season. Figure 3.1 shows the temperature and dewpoint temperature profiles for five selected days of IOP #1, before the onset of the monsoon. The soundings were made at 23 UTC, which corresponds to 4pm local time. The profiles show that the boundary layer height fluctuates between 800 and 710 mb and that the atmosphere is very dry. These observations were in agreement with the lack of low level cloudiness observed during this IOP.

Fig. 3.1. Temperature and Dewpoint profiles for 5 selected days during the IOP 1.

Figure 3.2 shows the thermodynamic profiles observed during the IOP #3. The observations were taken at 21 UTC (2 pm). They show that the boundary layer depth is between 850 and 770 mb. The atmosphere is also more moist and the closeness of the temperature and dewpoint profile curve near the top of the boundary layer agrees well with the observed presence of scattered cumulus clouds at this time during most of the days.

Figure 3.3 is the same as Fig. 3.2, except for five selected days from the IOP #4. The results show that the boundary layer depth is even more reduced with an average depth between 890 and 830 mb. The lower atmosphere is slightly drier in this period when compared to the IOP 3.

Fig. 3.2. Temperature and Dewpoint profiles for 5 selected days during the IOP 3.

Fig. 3.3. Temperature and Dewpoint profiles for 5 selected days during the IOP 4.

Figure 3.4 shows the evolution of the foliage during IOP # 2 for the point #1. Figure 3.5 displays the same information but for point #15.

Fig. 3.4. Evolution of the foliage during the IOP3, 1 km east of Tesopaco.

Fig. 3.5. Evolution of the foliage during the IOP3, 2 km southeast of Tesopaco.

4. Year two activities

Data processing

a) Pilot balloon data processing

During the coming months the pilot balloon data will be all placed in an electronic format and then fully quality controlled. This task will be carried out with the help of two undergraduate students, one of whom participated in the observations in Tezopaco. The design of the pilot balloon array will then allow us to calculate a variety of meteorological quantities over different triangles and polygons. The data will be used to identify the seasonal evolution of land-sea breezes, and infer the changes in the intensity of the land breezes, sea breezes and the patterns of vertical motion.

b) Rainfall data processing

The rainfall data stored in forms still needs to be collected by each municipality in Mexico and then gathered by Proteccion Civil from Sonora. When this task is completed, all the forms and information about the location of the gauges will be transferred to the NSSL for quality control and analysis. The information then needs to be introduced into an electronic format and further quality controlled.

The final rainfall dataset will complement other indirect measures of rainfall available such as the GOES visible and IR imagery and the Mexican Weather Service (SNM) radar data from the Guasave and Obregon radars.

We expect the rainfall data to be useful in identifying the date of onset of the rains for comparison with the vegetation and diurnal breeze changes. The stations will also be composited with respect to distance from the coast to see if the hypothesized mountain – to - coast rainfall propagation was clearly observed during the summer of 2004.

c) Additional Data

Additional available datasets include the radiosonde and tethersonde observations, hourly surface observations, cloud photography and vegetation photography. These data, together with the rainfall and pilot balloon information, will be quality-controlled and made available via the following website: http://www.nssl.noaa.gov/projects/pacs.

Research foci

The research activities during the next year will focus on identifying the seasonal changes in the diurnal circulations at all of the sounding sites and distinguishing the seasonal evolution from the more rapid change that might be expected with the onset of the rains. We expect to be able to develop a composite of the boundary layer structure and diurnal cycle at the Tezopaco site that represents pre- and post onset conditions.

One graduate student will begin at MS degree program in January 2005 at the University of Oklahoma. The student, from Mexico, spent two months at the Tezopaco site in August and September 2004, and is familiar with the data set collected. Another Mexican student will visit NSSL in January-March 2005, to work towards finishing a thesis required by student of the Bachelor’s degree program in Meteorology at the University of Veracruz in Xalapa, Mexico. This student also spent two months at the Tezopaco site.

Funds are being requested to continue operation of the flux tower for a second summer. These funds should cover the site’s operation and associated data processing and research activities at the University of Sonora under the supervision of the co-PI Chris Watts. A preliminary report showing some of the results from the flux tower observations is attached to this report.