BACKGROUND Goals of the Desert Winds Project The primary purpose of the Desert Winds Project is to obtain high resolution meteorological data and related surface geological and vegetation data for natural (e.g., uncultivated) desert sites where wind is or has been a major erosive or depositional force. The objectives are twofold: (1) to provide the detailed field measurements needed to carry out quantitative studies of wind as an agent of surface geologic change; and (2) to establish a baseline for defining the "normal" range of climatic conditions that can be expected to occur on a decadal time scale, in areas considered representative of the major American deserts. The Gold Spring locality was selected to represent that part of the Great Basin Desert that extends into northeastern Arizona. The long-term goal for acquiring and analyzing the Desert Winds Project data is to use them to address problems of land resource degradation by wind, whether resulting from climatic variation (aridification) or human activities (desertification), or both. Location of the Geomet Station The data on this disk were obtained by instruments deployed on an automated, solar-powered, satellite-relay station, operated by the U.S.Geological Survey Desert Winds Project at lat 35 degrees 46' 45" N, long 111 degrees 03' 18"W (T27N, R12E, NE 1/4 of Sec. 4) in northeastern Arizona. The Geomet station (where both geological and meteorological data are collected) was deployed at Gold Spring on August 2, 1979, on Navajo and Hopi Indian reservation land, in an area considered one of the most severely desertified in the U.S. (Dregne, in Sheridan, 1981). Climate The Gold Spring area is designated as arid steppe (Hendricks, 1985). The Geomet station there has recorded temperature extremes that range from 38.2 degrees C in summer (1989) to -25.3 degrees C in winter (1982). Maximum peak windgusts were recorded at 30.4 m/s (68.1 miles per hour) in 1984 (Helm and Breed, unpublished information). Average annual precipitation at the station during the 1980-1992 period of record is 142.5 mm (this figure includes both summer rain and winter snow, but because no special equipment is installed to measure snowfall, the precipitation gauge probably underestimates the available winter moisture). Geology The geologic setting of the Gold Spring station is representative of much of the plateau and canyon country in the southern part of the Colorado Plateau Province. The station sits on the surface of the Moenkopi Plateau, at an elevation of about 1,667 meters above sea level. The Moenkopi Plateau is underlain by sedimentary rocks (mostly sandstones) of Mesozoic age that dip very gently (1 degree to 2 degrees) to the northeast. The Geomet site lies 2 km northeast of 158-m-high cliffs that mark the edge of the Moenkopi Plateau and which topographically separate the upland from the badlands east of the Little Colorado River, below. Total relief from the top of the plateau to the river floodplain is about 300 meters. The cliff edge has been intricately dissected by episodic (seasonal) fluvial runoff and sapping into many reentrants, of which Gold Spring is one of the most prominent. Wind erosion has modified the reentrants to U-shaped amphitheaters, from which the covering sand dunes have been stripped and in which parts of the exposed bedrock have been streamlined into yardangs (Breed and others, 1984). Surface Features The surface of the Moenkopi Plateau is a gently rolling sand plain developed on an unconsolidated eolian sand sheet and crossed by subparallel sand ridges (linear dunes) that extend northeastward from the cliff edge. The dunes range in age from Pleistocene or older to late Holocene (Breed and Breed, 1979; Breed ond others, 1984; Billingsley, 1987a,b; Stokes and Breed, 1993). The Geomet station is on a slight rise in an interdune corridor. Local topographic relief within the one- square-mile area surrounding the Gold Spring station is less than three meters. A geologic map (scale 1:31,680) of the area encompassing five 7 1/2-minute quadrangles in the vicinity of the Gold Spring station shows details of the dunes and other surficial units in relation to the bedrock outcrops (Billingsley, 1987a). Soils Soils developed in the sand cover around the Gold Spring station are fine-grained sandy loams. They are mapped as mesic arid soils belonging to the Fruitland-Camborthids-Torrifluvents Association (Hendricks, 1985), although the parent material in the Gold Spring area is mostly of eolian, not alluvial origin. The average (modal) grain size of the sand sheet at the station is 0.104 mm (very fine sand). The sand sheet is partly mantled by a lag gravel of angular chert fragments a few cm to about 3 cm in diameter, which are let down by erosion from the local bedrock outcrops. Trenching in the interdune corridor near the Geomet station shows 1.1 m sand with a weak soil over weathered Navajo Sandstone (Tuesink, unpublished information, 1993). Vegetation Vegetation at the Gold Spring site is typical Great Basin plains grassland, surrounded on the rest of the Moenkopi Plateau by Great Basin desert scrub (Hendricks, 1985). There are no trees. The vegetation community in the interdune corridor near the Geomet station is dominated by broom snakeweed and galleta, a grass that grows as individual tufts; nearby dune crests are dominated by Ephedra (Mormon tea) and by sandhill muhly, a grass that grows in mats. Most other vegetation around the Geomet station consists of yucca, black grama, and needle-and-thread grass (Musick, unpublished information). Land use The Gold Spring area is not farmed, but is used for grazing of cattle, sheep and horses by local residents, members of the Navajo Indian Nation who have joint use of this land with the Hopi Indian Tribe. To protect the Geomet station from grazing animals, a three-strand barbed wire fence was constructed in 1979 to enclose the immediate area around the basic station. The fenced area was enlarged in June 1990 to accommodate the additional towers of the superstation. ___________________________________________________________________________ Gold Spring Site Hardware Descriptions The stations used by the Desert Winds project are designed to provide field-based physical data of sufficient resolution to support scientific studies of surface (mostly eolian) processes under monitored climatic conditions. Because they collect both geological and meteorological data, we refer to them as Geomet stations. Located in remote areas, the stations are automated and solar powered. They are unattended on a daily basis, but they do require scheduled servicing and repair: weekly cleaning and checking by a custodian and periodic servicing (including calibration) and repair, generally at 6- month intervals, by a skilled technician (table 1). At 1- or 2-hour intervals each station transmits digital data from its array of sensors to the Geostationary Operational Environmental Satellite (GOES). From GOES, the data are relayed to (and stored at) the receiving station of the National Environmental Satellite Data Information Service (NESDIS) at Wallops Island, Virginia; NESDIS is operated by the National Oceanic and Atmospheric Administration. From Wallops Island, the data are transmitted via land lines to the U.S. Geological Survey computers at project headquarters in Flagstaff, Arizona. Table 1 - Scheduled maintenance at GEOMET stations _________________________________________________________________ Operation Frequency Performed by _________________________________________________________________ Reset data-collection 6 months Technician platform clock Rebuild/calibrate As necessary Technician anemometers Rebuild/calibrate As necessary Technician wind-direction sensor Clean air-temperature/ 6 months Technician relative-humidity filters Clean solar panels Weekly Custodian Clean precipitation As necessary Technician gauge or Custodian Check radio and data- 6 months Technician collection platform Collect samples from Weekly Custodian windblown-particle catcher Check power systems 6 months Technician Replace power-system 2 years Technician batteries Check all temperature 6 months Technician sensors for accuracy Check relative humidity 6 months Technician against psychrometer _________________________________________________________________ Design of the Gold Spring Geomet Station The original Gold Spring station was installed in 1979. This was a basic station, equipped with standard sensors to record observations of a set of near-surface phenomena: average and peak-gust wind speeds, wind direction, precipitation, relative humidity, barometric pressure, air temperature, and soil temperature (table 2). The Gold Spring site was retrofitted to superstation status in 1990. The superstation has a much larger and more varied complement of sensors than the basic station (table 3). The current Geomet station at Gold Spring consists of two free- standing towers, a solar power system, a data-collection platform (DCP) with a radio transmitter, and nineteen sensors that provide twenty-two parameters. The DCP tower has a base tripod 1.8 m on a side and 1.6 m tall, with a 6.1-m mast mounted on one corner. It is constructed of high-quality aluminum pipe. The wide base makes this tower very stable, even in peak-gust winds of nearly 80 knots. Three adjustable footpads enable the tower to be leveled, and rebar stakes driven through holes in the footpads increase its stability. A lightning rod on the top of the mast is connected with a ground rod at the base of the tower. A hinge arrangement at the bottom of the mast allows it to be lowered for servicing of instruments. Several sensors mounted on the tower are connected with a DCP attached to one side of the base tripod. A crossarm at the top of the mast provides a mounting surface for the wind-direction sensor. The basic station had a single wind-speed sensor also mounted on the crossarm, whereas the superstation has three wind-speed sensors mounted on a separate wind tower (described below). A precipitation gauge is fastened to the DCP tower on top of the tripod. The barometric-pressure sensor is mounted to the base 1.2 m above the ground. The soil-temperature probe is buried 12.7 cm (basic station) or 4 cm (superstation)beneath the ground surface, outside the perimeter of the DCP tower. The basic station had one relative- humidity and one air-temperature sensor mounted to the DCP tower base at the 1.6-m level. The superstation has one relative-humidity and two air-temperature sensors mounted on the wind tower (described below). The superstation has, in addition to the DCP tower, a wind tower. This structure is a 6-m tall, three-sided (31.75 cm on a side) communication tower. A lightning rod attached to the top connects with a ground rod at the base of the tower. Three guy wires provide stability: two are attached to earth anchors, the third, to a winch that allows the tower to be lowered. Lowering, easily accomplished by one person, allows servicing of tower instruments from the ground. Wind-speed sensors mounted on crossarms at 1.2, 2.7, and 6.1 m above the ground provide profiles of the average and peak-gust windspeeds, from which key parameters are calculated for studies of wind erosion. Air-temperature sensors are mounted on the crossarms of the wind tower at 1.2 and 6.1 m above the ground; the lower sensor is coupled with a relative-humidity sensor. Two buried soil-temperature probes at 10- and 20-cm depths have been added to the 4-cm probe to provide a profile of soil temperatures. An experimental automated sensor (SENSIT) was deployed at Gold Spring on February 21, 1991; it records the impacts of saltating sand grains on a piezoelectric crystal mounted 5 cm above the ground. The SENSIT is not a standard, "off-the-shelf" sensor, and the data it provides remain experimental. Because these data require further study and interpretation by Project scientists they are not included in this archival dataset. The Geomet station has three power requirements: power for the DCP electronics and GOES transmitter; power for the basic sensors; and, with the superstation, power for additional sensors. The station is in a remote location away from commercial power, and planned maintenance visits to the site is at 6-month intervals (table 2). The power systems must, therefore, be self sufficient, and cannot require constant attention. The station uses battery power (gel-cell type) and solar panels, both of which require little maintenance. Solar panels supply daytime power and charge the batteries. The choice of this system imposes power restrictions on the DCP, and all sensors must have low power requirements. The basic station DCP contains one battery charged through an internal regulator by an external solar panel. An additional input allows use of external power. Internal circuitry and program control allow the sensor power to be turned on just prior to and during a sampling period. Power to the basic sensors is turned off at all other times. Such a switched power provision is important where power is limited. The superstation requires additional power to operate the SENSIT sand-flux sensor which requires constant power. The necessary power is supplied by additional batteries charged by two 40-watt solar panels. The DCP is the heart of the station. The DCP is a specially designed Handar 540A-1 (Handar, 1990) enclosed in a box 40 cm wide, 30 cm high, and 23.75 cm deep. The DCP box contains a programmable microprocessor, memory for program and data storage, and as many as six sensor interface cards, and a GOES radio transmitter. All sensors are interfaced to the 540A-1 DCP through a junction box, also mounted on the DCP base tripod. The solar-powered battery system, coupled with radio transmission via Earth-orbiting satellite, allows operation of Geomet stations in remote areas, and it obviates the need to visit the stations frequently to collect data recorded on paper rolls or digital tapes. (Local custodians do visit the sites weekly to protect it from vandalism, collect sediment from the windblown-particle catcher and to remove dust from solar panels). Automated DCPs generally work very well--the Geomet station at Gold Spring has operated almost continuously, without major problems, since August 1979. Problems with sensor performance, station operations, or satellite transmission are usually immediately apparent as gaps or errors in the electronic data stream received in Flagstaff. Table 2 - Range, precision and sample interval of basic-station sensors _____________________________________________________________________________ SENSOR RANGE ACCURACY _____________________________________________________________________________ Sampled at 30-minute intervals - (October 27, 1979 through April 24, 1980) Sampled at 6-minute intervals - (April 25, 1980 through August 8, 1990) _____________________________________________________________________________ Wind Direction 0-360 DEG. +-5.5 degrees to 9.2% Wind Speed 0-100 MPH +-2.5% to 5.0% Peak Gust 0-100 MPH +-2.5% to 5.0% Precipitation 0.00-99.9 +-.01" Soil and Air -30 deg. C +-1-2% Temperature to 70 deg. C Relative 0-100% +-3-5% Humidity Barometric Dependent on +-1-1.4% Pressure elevation ____________________________________________________________________________ Table 3 - Range, precision and sample interval of super-station sensors _____________________________________________________________________________ SENSOR RANGE ACCURACY _____________________________________________________________________________ August 9, 1990 through DECEMBER 31, 1992 ___________________________________________________________________________ Sampled at 6-minute intervals _______________________________ Wind Direction 0-359.9 deg. +- 5% Wind Speed 0-150 MPH +- 5% Peak Gust 0-150 MPH +- 5% SENSIT Sand Flux experimental Sampled at 12-minute intervals _______________________________ Precipitation 00.0 - 99.9 +- 0.3% Air Temperature -50 to 50 +- 0.2 degrees,0-60 deg.C degrees C +- 0.6 degrees,-50-0 deg.C Sampled at 1-hour intervals ____________________________ Relative Humidity 0-100% +- 2%,0-80% RH Soil Temperature -50 to 50 +-.1 deg.C,-20 to 50 deg.C degrees C +-.6 deg.C,-50 to -20 deg.C Barometric Pressure Dependent +- 0.3 % on altitude of site. Battery voltage 0 - 99.9 +- 5% ___________________________________________________________________________ INSTRUMENTATION Following are detailed descriptions of the instruments deployed at the Gold Spring Geomet station and their requirements for calibration and maintenance (tables 1, 2 & 3). Data-collection platform (DCP) Description: The DCP processes and stores the signals from the sensors until the data are transmitted via the GOES satellite. The transmitting antenna to the satellite is a cross-element yagi attached to one corner of the DCP tower. The GOES satellite radio assembly transmits approximately 10 watts of power on a frequency assigned by the National Environmental Satellite Data Information Service (NESDIS). Calibration: All DCP circuits are checked on each maintenance visit to the Geomet sites. The visits, at 6-month intervals (table 1), are made by a qualified electronics technician. The circuit checks require insertion of the proper stimulus and measurement of the result for each circuit. The DCP manufacturer markets a pluggable card for this purpose, which is useful in the laboratory. In the field, however, we have found that hand-held simulators (handmade in our laboratory) are more useful, because they allow introduction of electronic stimuli so that cables and connectors can be checked as well. A simple switch provides the necessary stimulus to check the precipitation interface circuit. A precision resistance simulates a given temperature and is used to check each of the temperature interface circuits. A hand-held voltage source provides the stimulus to calibrate the relative- humidity circuitry. Another hand-held device generates a pulse to check wind-speed circuits. The technician also measures output power from the GOES radio at both the radio and the antenna, and reflected power at the antenna. Maintenance: Maintenance includes examination of all cables and connectors for proper connections, cleanliness, and damage. A desiccant (moisture-absorbing crystals) inside the DCP enclosure box effectively removes moisture from the box, thus preventing damage to the electronics. The desiccant is replaced each time the DCP is opened; used desiccant is baked for 12 to 24 hours, stored in an airtight container, and reused. Air-temperature sensor Description: This instrument is a solid-state precision linear thermistor placed in a compact cylindrical assembly. The resistance of the thermistor is inversely proportional to the temperature. The resistance is measured by a voltage divider circuit, and the result is linearized by a look-up-table routine. A convection-aspirated shield around the thermistor reduces the effect of solar radiation on air temperature and protects the thermistor from precipitation. This instrument is mounted either to the DCP tower at the 1.8 m-level (basic station), or to the wind tower at the 1.2- and 6.1-m levels (superstation). It is simple, rugged, and reliable. Calibration: Calibration requires taking a temperature reading near the thermistor with a hand-held thermometer. A chart provides a value for the resistance, which must agree with the resistance of the instrument, taken with an ohmmeter. Cross-checking the temperature readings from the two sensors (1.2- and 6.1-m) at the superstations provides a daily indication of their performance. Maintenance: Periodic maintenance consists of removing the shield and cleaning a filter that protects the thermistor. Relative-humidity/air-temperature Description: The temperature-sensor part of this combined instrument is the same as the air-temperature sensor described above. The humidity sensor is a capacitor consisting of a metallic grid deposited on a glass plate and covered by a hygroscopic polymer film. The dielectric constant of the polymer material increases with the amount of water absorbed, such that capacitance depends on the relative humidity. The basic response to changes in humidity is rapid (approximately 1 second to reach 90 per cent of the final value). However, a filter that protects the sensor from dust and physical damage slows the response to 20 to 30 seconds. The humidity-recording part of this sensor is mounted next to the air-temperature thermistor, so that both are sampling the same volume of air. The instrument is mounted on the base of the DCP tower (basic station) or at the 1.2-m level of the wind tower (superstation). Calibration: Calibration of the humidity sensor requires a strictly controlled laboratory environment. Therefore, when calibration is necessary, the instrument is returned to the manufacturer's facility. The manufacturer suggests calibration every 24 months. At 6-month intervals the instrument's accuracy is checked by comparing its readings with those taken by a psychrometer. The psychrometer consists of two thermometers that are identical except that the bulb of one is wrapped with a wet wick. Evaporation from the wet bulb causes it to register a lower temperature than that of the dry bulb. The difference between the two temperatures constitutes the relative humidity. Maintenance: Periodic servicing consists of removing the shield and cleaning a filter that protects the capacitor. Soil-temperature sensor Description: This sensor is a precision linear thermistor housed in a waterproof, stainless steel enclosure. The resistance of the thermistor is inversely proportional to the soil temperature. Resistance is measured by a voltage divider circuit, and the result is linearized by a look-up-table routine. The probes lie in the soil at depths of 4, 10, and 20 cm, and they are connected to the DCP by weather- and animal-proof cables. Calibration: A hand-held thermometer is used to check the soil temperature near the probes. This reading is cross referenced to a resistance chart. If calibration is correct, the resistance value of the chart will match the resistance of the instrument as measured with an ohmmeter. Faulty instruments are replaced and then repaired in the laboratory. Daily monitoring and cross-checking of data transmitted from each of the three probes generally indicates their condition and accuracy. When working properly, the three probes follow the same basic pattern of response, with increasing lag and decreasing extremes as depth increases. Maintenance: These sensors are reliable, and they require little maintenance other than checking their depth. When installing them, we insert a surveyors' flag at the 4-cm depth and mark the flagstaff's position at the ground surface. This mark is checked during each visit to a site; if it is unchanged, we know that the instruments have remained at the required depth. To mark the flagstaff's position, we have found that a wheel collar used to hold wheels on model airplanes works well--it does not erode away as pen marks do or move as tape does. Some time is needed for the ground to stabilize after sensor installation, and we therefore try not to cause any unnecessary disturbance once the instruments have settled in. Barometric-pressure sensor Description: The barometric-pressure sensor is an aneroid barometer that utilizes an evacuated bellows sensitive to changes in absolute pressure. As pressure changes, motion of the bellows moves a contact across a precision potentiometer, producing an output resistance proportional to the barometric pressure. Housed in a weather-proof container, the barometer is mounted on the DCP tower base tripod. Calibration: If the sensor fails or records spurious readings it is sent to the manufacturer for repair and/or calibration. Maintenance: Insuring that the holes that allow atmosphere to the bellows are clean and that the instrument remains upright are the only maintenance requirements. Precipitation sensor Description: The precipitation sensor is a funnel with a 20-cm-diameter orifice to collect and direct rain to a tipping bucket mechanism. The funnel and mechanism are housed in a metal cylinder that is secured to a corner of the DCP tower, 1.8 m above the ground. Coupled to the tipping bucket is a magnetically activated reed switch, which momentarily closes each time the bucket tips; 0.25 mm of collected water causes the bucket to tip. The water drains out the bottom of the gauge and a second bucket moves into position to collect more water; the procedure is repeated until no more is available for collection. Switch closures are counted by an interface circuit in the DCP. This hardware count is periodically added to a cumulative count maintained by DCP software. Calibration: Slowly pouring measured amounts of water through a clean, level instrument will verify calibration. Enough liquid is poured through the instrument to cause 10 tips. This measured amount is then divided by 10 to produce an average amount per tip. Mechanical adjustments, if necessary, are then made. Maintenance: Maintenance consists of cleaning debris from inlet and outlet screens and from the tipping mechanism and keeping the instrument level. Wind-direction sensor Description: This instrument is a heavy-duty aluminum and stainless steel wind vane that is coupled to a precision micro-torque potentiometer, which has low rotational torque. The wire-wound potentiometer produces an output resistance that varies in proportion to the wind direction. The wind direction vane is mounted on the DCP tower at the top of the 6.1-m mast. Calibration: Calibration is accomplished by inserting a locating pin into the body of the instrument to orient the wind vane with an internal potentiometer. Pointing the vane exactly to the south and locking the instrument in place completes the calibration. Maintenance: Maintenance requires observation of the instrument for signs of damage and verification that readings (in degrees) agree with vane direction. This sensor is very reliable and normally requires little maintenance except for protection from birds. We do this by attaching verticle darning needles to the vane. Wind-speed sensor (anemometer) Description: The anemometer consists of three cups mounted on a rotor shaft that rotates on instrument-quality stainless steel bearings. A magnet on the rotor shaft moves near a reed switch (basic station) or a Hall Effect transducer (superstation), which detects the field of the rotating magnet. This detected field produces a pulse rate in proportion to the wind velocity. A single anemometer is mounted to the top of the DCP tower mast at the basic station. Three anemometers are mounted on the wind tower at each superstation, at 1.2, 2.7, and 6.1 m above the ground. Calibration: Operations of the anemometers are checked each time they are rebuilt. In 1993 we began calibration in a wind tunnel, which produces a calibration constant for each instrument. This constant is entered into the DCP program upon installation of the instrument. When placed on a tower in the field, anemometers are considered to be working properly if the daily average of anemometer readings at each of three heights plots as a straight line on a log-height, linear average-wind-speed graph. This field method, however, only insures consistent operations between anemometers and is not a true calibration against a known standard. At present, anomalies in the wind-speed data are common, and it is difficult to distinguish signal from noise. Maintenance: The anemometers require periodic cleaning and bearing replacement. The period between their rebuilding depends on the frequency of storms and the type of airborne contaminants (such as dust and salt) that get into the bearings. Daily checks, made by cross-checking the three wind speeds transmitted from each station, provide a good indication of the condition of the anemometers. Windblown-particle catcher Description: A field dust sampler, designed for U.S. Department of Agriculture (USDA) research (Fryrear, 1986), collects samples of sand and dust particles moving at three levels above the ground. The Desert Winds standard heights for the top two sample pans are 50 and 100 cm, with the lowest sampler pan set as closely as possible to the 5-cm height of the sand-flux sensor. A wind vane on each sampler pan keeps it pointed into the wind. Particle-laden air passes through each sampler opening; once inside, it reduces speed and particles settle in the sampler pan. (For an analysis of sampler performance, see Stout and Fryrear, 1989.) Because the data are used for experimental testing, they are not included in this dataset. Calibration: Calibration consists of checking and adjusting the heights of the three catchers. Maintenance: Maintenance of this mechanical device requires cleaning of plugged screens, insuring free rotational movement of the catchers, tightening screws, and replacing worn parts. Sand-flux sensor Description: The sand-flux sensor is an experimental device that automatically detects particles moving in saltation (bouncing along the ground). The sensor has no moving parts and is invariant to wind direction (which is recorded independently by the Geomet wind-direction sensor). The upper part of the SENSIT contains an exposed, ring-shaped, piezoelectric crystal, which is mounted on a post connected to a buried electronics housing. Windborne grains (in the diameter range of about 50 um to 1 mm) impact the sensor, deform the piezoelectric crystal, and produce a charge proportional to the impact. The resultant signals represent particle movement; each pulse represents one count. At present, a hardware limitation enables us to record only one of these signals, as a frequency. This deficiency limits our use of the SENSIT to the detection, but not yet to the measurement, of particle movement. Because the SENSIT data are experimental, they are not included in this dataset. Data Collection, Storage and Retrieval Data collected from the automated sensors at each Geomet site must be directed through a series of steps within the DCP before becoming recognizable and usable for scientific research. These steps are directed by a software program within the DCP. Generally, one software "channel" for each sensor controls the program actions required to monitor it. However, for the wind-speed sensors, two software channels are needed for each hardware sensor; one channel monitors average wind speed while the other monitors peak gusts. The user programs each channel by specifying the monitoring schedule, the data-processing algorithms, the location of the hardware sensor input to be monitored, and the format in which the data are to be transmitted. Once programmed, the DCP continuously samples data from each sensor on the programmed schedule. The original data signals are produced by the sensors in various forms: some sensors produce analog voltages, others vary a resistance, and still others produce electrical pulses. Each signal is converted in real time into digital form by interface circuit cards. Each channel's data- processing algorithm then converts this digital data into its final form (average, peak, temperature, count). Lastly, the data are stored in the DCP memory, where they remain until transmitted on the GOES radio. The GOES satellites are in geostationary orbit above the Earth and are nominally stationed 23,000 miles (37,000 km) above the equator. The two available for our use are GOES East, at approximately long 75 deg. W., and GOES West, at approximately long 135 deg. W. At present, the Gold Spring Geomet station is assigned a channel and time slot on GOES West. Once an hour the DCP program collects Geomet data from memory and formats them so that they can be transmitted in pseudo-binary form, a compression code that enables transmission within the time frame allotted by NESDIS. Transmission to GOES is in the 401.7- to 402.0-Mhz band. GOES West receives this signal and retransmits it as an 1694.5-MHz S-band signal to a receiving station at Wallops Island, Virginia. At Wallops Island, the Data Collection System Automatic Processing Subsystem sorts and stores the data in user files. High-speed dedicated or low-speed dial-up lines transfer the data to Desert Winds project headquarters in Flagstaff, where software performs a series of checks to document missing or erroneous data. Finally, software converts the pseudo-binary code to ASCII, separates it by station and time, formats, and stores the data on disks and tapes for analysis by scientists. Detailed coding of the data for quality, changes in sensors, and gaps due to NESDIS or Geomet station malfunctions is necessary to prepare the data for publication (table 4). ___________________________________________________________________________ Table 4. CODE VALUES FOR: SENSOR SENSOR PARAMETER DOWN NOT INSTALLED WIND DIRECTION -99.0 N/A WIND SPEED (6.1m) -99.0 N/A PEAK GUST (6.1m) -99.0 N/A WIND SPEED (2.7m) -99.0 -77.0 WIND SPEED (1.2m) -99.0 -77.0 PEAK GUST (1.2m) -99.0 -77.0 PRECIPITATION -99.0 N/A AIR TEMPERATURE (6.1m) -99. -77. AIR TEMPERATURE (1.2m) -99. N/A RELATIVE HUMIDITY -99. N/A BAROMETRIC PRESSURE -99. N/A SOIL TEMPERATURE ( 4cm) -99. -77. SOIL TEMPERATURE (10cm) -99. -77. SOIL TEMPERATURE (20cm) -99. -77. __________________________________________________________________________ REFERENCES Billingsley, G.H., 1987a, Geologic map of the southwestern Moenkopi Plateau and southern Ward Terrace, Coconino County, Arizona: U.S. Geological Survey Misc. Inv. Map I-1793, scale 1:31,680. Billingsley, G.H., 1987b, Geology and geomorphology of the southwestern Moenkopi Plateau and southern Ward Terrace, Arizona: U.S. Geological Survey Bulletin 1672, 18 p. Breed, C.S., and Breed, W.J., 1979, Windforms of central Australia, and a comparison with some linear dunes on the Moenkopi Plateau, Arizona: in El-Baz, F. (ed.), Scientific results of the Apollo-Soyuz Mission: National Aeronautics and Space Administration (NASA) Special Paper 412, p. 319-358. Breed, C.S., McCauley, J.F., Breed, W.J., McCauley, C.K., and Cotera, A.S., 1984, Eolian (wind-formed) landscapes: In Smiley, T.L., Nations, J.D., Pewe, T.L., and Schafer, J.P. (eds.), Landscapes of Arizona- The Geological Story. University Press of America, Lanham, MD, p. 359-413. Dregne, Harold, 1977, Desertification of arid lands: Economic Geography, v. 53, no. 4, p. 325. Helm, P.J., and Breed, C.S., unpublished data, Instrumented field studies of sediment transport by winds: In Breed, C.S. (ed.), Monitoring wind-related surface processes at desert sites in Arizona, New Mexico, and California: USGS Bulletin (in preparation, 1995). Hendricks, D.M., 1985, Arizona soils: College of Agriculture Centennial Publication, Univ. of Arizona, Tucson, 244 p. McCauley, J.F., Breed, C.S., Helm, P.J., Billingsley, G.H., MacKinnon, D.J., Grolier, M.J., and McCauley, C.K., 1984, Remote monitoring of processes that shape desert surfaces: The Desert Winds Project: U.S. Geological Survey Bulletin 1634, 19 p. Sheridan, David, 1981, Desertification of the United States: Council on Environmental Quality, U.S. Government Printing Office, Washington, D.C., 142 p. Stokes, Stephen, and Breed, C.S., 1993, A Chronostratigraphic re-evaluation of the Tusayan Dunes, Moenkopi Plateau and southern Ward Terrace, northern Arizona: in Pye, K. (ed.), The dynamics and environmental context of aeolian sedimentary systems: Geological Society (London) Special Publication no. 72, p. 75-90. Handar, 1990, Handar model 540A Data Acquisition System Operation and Service Manual, 1288 Reamwood Avenue, Sunnyvale, CA. 94089 Handar, 1990, 1989-1990 Handar Product Catalog, 1288 Reamwood Avenue, Sunnyvale CA. 94089