The Cumulus Photogrammetric, In-situ and Doppler Observations (CuPIDO) field experiment will use digital visible spectrum cameras, surface mesonet stations, high temporal resolution soundings and aircraft data to examine the onset and transition of shallow to deep orographic cumulus associated with the North American Monsoon (NAM) in southern Arizona. CuPIDO will took place from July to August of 2006, over the Santa Catalina Mountains north of Tucson. The aircraft observations, and airborne radar data, are intended to describe the cloud microphysical and thermodynamic structure of cumuli and the mesoscale environment in which they build and decay.

CuPIDO is designed to study the development of thunderstorms that form during the summer monsoon season. CuPIDO will consist of observations taken from sets of digital cameras, six 10 meter portable automated surface weather stations, four 10 meter tower-based integrated surface flux facilities (ISFF), two balloon sounding systems (MGAUS), the Wyoming King Air research aircraft carrying a W-band dual-beam radar and the University of Alabama, Huntsville MIPS. MGAUS, WKA and MIPS will operate from 17 Jul - 18 Aug, 2006

The purpose of CuPIDO is to examine the development of the earliest stages of the thunderstorms. The clouds begin to form over the highest peaks each morning and the reliability of the formation makes these areas natural cloud laboratories. In contrast to conventional "storm chasing," where investigators intercept the weather systems, CuPIDO will measure conditions prior to the storms beginning and monitor the way that the initial, shallow clouds interact with and modify the environment as they develop. The information gained will help improve the forecast models used by the National Weather Service and ultimately provide more accurate and timely warnings.


CuPIDO Links


Site Survey Information

  • September 5th
    • Photos
    • Notes from Milizer
    • Notes from Horst
  • January 06th
    • Photos
    • Notes from Poulos
  • Photos
  • Desert Life & Landscape
Sep 05 Photos Notes: Militzer (text) Notes: Horst (text)
Jan 06 Photos Notes: Poulos (Word)
Photos Site Photos Desert Life&Landscape AirSpaceMuseum



See schedule and logistics here.


CuPIDO Measurement Sites

A Map of all ISFF PAM and ISFF tower site locations.

Click on the site number to get a more specific description of a given site

Station # Station ID and Name Latitude
(deg N)
(deg W)
m (feet)
1 N - North - Campo Bonito 32° 34.303' 110° 42.401' 1305 (4362)
2 NE - Northeast - Stratton Canyon 32° 30.243' 110° 40.772' 1365 (4560)
3 ENE - East Northeast - Davis Mesa 32° 27.855' 110° 37.400' 1234 (4120)
4 E - East - Lone Hill 32° 24.716' 110° 35.517' 1141 (3812)
5 SSE - South-southeast - Bellota Ranch 32° 18.200' 110° 36.709' 1270 (4244)
6 S - South - Bug Springs 32° 20.440' 110° 42.897' 1500 (5011)
7 SW - Southwest - Pusch Ridge, Archery site 32° 21.328' 110° 57.526' 840 (2805)
8 W - West - Catalina 32° 25.928' 110° 55.331' 853 (2852)
9 WNW - West Northwest - Golder Ranch 32° 28.711' 110° 53.026' 961 (3210)
10 NNW - North Northwest - Rancho Solano 32° 33.661' 110° 51.172' 1044 (3490)
Site order by elevation is: 7, 8, 9, 10, 4, 3, 5, 1, 2, & 6


Solar Characteristics for CuPIDO Location


  • Layout of the Basic towers (N, E, S, SW, W, WNW and NW) in PDF format.
  • Layout of the ENE and SSE flux towers with 180 degree sonic orientation PDF format.
  • Layout of the NE flux tower with 135 degree sonic orientation PDF format.
  • Layout of the NNW flux tower with 210 degree sonic orientation PDF format.

A total of 128 meteorological variables will be recorded by the ISFF team on behalf of CuPIDO's scientific goals. Each of the 10 tower sites is 10 m in height and they are outfitted as follows:

  • All 10 towers have a basic set of meteorological instrumentation: 1 Hz TRH, wind direction, wind speed, barometric pressure, net radiation, and a tipping-bucket precipitation gauge. TRH, net radiation, and P are at 2 m agl while the anemometer is at 10 m agl. Each site will have GOES satellite communication of 5 minute average quantities if the GOES satellite is available. See pressure measurement references for the Vaisala PTB220B, barometer and the single-disk static pressure port.
  • The four flux tower (ISFF) sites (NE, ENE, SSE and NNW) will each be outfitted with a single Campbell CSAT3 thermosonic anemometer at 7 m (mean winds, momentum and sensible heat fluxes); a Campbell KH2O Krypton hygrometer at 7 m (water vapor fluxes), located ~25 cm behind each of the CSAT3s. Each flux site also has a "darkhorse" radiometer stand with 4-component radiometers of type Epply PSP and PIR. Located near this stand are the soil sensors: heat flux (5 cm), soil temperature (2-8 cm), soil moisture (5 cm), and soil properties (5 cm). In addition a LiCor model 7500 which measures fast-response CO2 and H20 will be placed at the SSE site (which requires the use, due to its weight, of a Rohn tower, rather than a PAM mast.


Data Collection Notes

This is the 2nd use of the Viper-based PC104 data systems (NIDAS, from which the HIAPER data system is derived), running the new data acquisition code for primary data acquisition. This is the first deployment of NIDS with the new GOES satellite transmit channel, 5 minute average data creation and near-real-time transmission. Online data plots will be available (dependent on GOES transmissions) and updated in ~30 minute increments. This is the first time 10 towers have been deployed with the NIDAS data acquisition system.

  • The ISFF base trailer will be located ~ 9 miles NW (and 20 minutes) of the Residence Inn at the University of Arizona Agricultural Research Station
    The UA Agricultural Research Station at 4101 N. Campbell Ave., Tucson, and location of the ISFF Base Trailer. The base trailer is located on the west side of the large white buildings near the center of the image (enter from the south gate).


Sensor Notes

General note

All data coming from sensors that were on the main mast will be removed from the NetCDF files while the mast was down or in the case of site 5, when there was an individual on the tower.


Sonic Anemometer

The sonic anemometers generally worked properly. Tilt corrections were applied to stations 2, 3, and 10, and station 5 was left untouched.

There were the usual bad values expected when rain affected the sensor, which is characterized when a drop of rain covers one of the 0.5 cm diameter transducers. Although no data was removed from the NetCDF files for the rain cases, very good indicators of quality of the data during these rain events can be viewed by plotting either uflag.7m, vflag.7m, wflag.7m, and tcflag.7m, which are fractions of the points that are detected as spikes. Also, one can plot diag.7m, which indicates the sonic detected a problem. Click on this link: sonic manual and look at pages 11 & 12 for a more detailed description. Lastly, one can plot samples.sonic, which indicates the amount of samples recorded in a 30 minute period, which should be 18,000 points. Anything less may indicate a problem associated with rain.

Please use the "Height.Site" link to view the tilt plots

Height.Site Boom Angle Actual Height (m) Lean (deg) Lean Azimuth (deg) 135° 7.0 2.9 -179.3
7m.ene 180° 7.0 1.7 -48.3
7m.sse 180° 7.0 0.2 -34.5
7m.nnw 210° 7.0 2.0 -78.3

All sonics will report at 60 sps to minimize the influence of pulse-sequence delay errors on the 20Hz fluxes.


Krypton Hygromter

The krypton hygrometers generally worked properly. Any spikes in the data associated with cleaning the sensor will be removed from the NetCDF files. During rain events, the sensor voltage was too low to be reliable, so those data will be removed from the NetCDF files too. There are also a few obvious outlying points that will be removed, reasoning for those outliers has not been established.

Values of sensible and latent heat flux derived from the measured sonic w, virtual temperature, and krypton hygrometer will be added to the NetCDF files.


Infrared Gas Analyzer, LICOR-7500

The open path licor generally worked properly other than expected bad values during rain events. Removing co2 data when the krypton hygrometer had a voltage reading at or below 0.01 V removed the vast majority of the rain cases. One other unknown outlier was removed too. We have yet to implement a Webb correction to remove humidity artifacts.


Prop-vane Anemometer

The prop-vane anemometers generally worked properly. There was in issue at station 3, where the starting threshold was too high because of a problem with a tight bearing. The issue seemed to have surfaced in mid July and got progressively worse until the sensor was replaced at 16:25 MST on August 9th, 2006. Because of that problem, the user should be cautious in using low windspeed values from the prop-vane anemometer during that period. One option is to use the values coming from the sonic anemometer at station 3. There are still some outstanding issues concerning boom angles, so the quality control process is not yet complete.


Temperature & Relative Humidity

The sensor generally worked properly. There were two instances where the sensor was pegged to full scale possibly because the A/D got stuck somehow, so that data will be removed from the NetCDF files. They occured during the following periods:


Surface Temperature

The sensor generally worked properly. There were a few instances (~4 total from 4 flux towers) when rain or unknown reasons caused unrealistic values, those values will be removed from the NetCDF files.

Note: The surface temperature sensor at Station 2 was aimed away from a large mesquite bush to a more representative spot on the ground at ~15:45 on July 14th.



The sensor worked properly.


Rain Gauge

The rain gauge generally worked properly. False rain tips were performed throughout the project to insure that the sensor was working properly, these false rain tips will be removed from the NetCDF files.



The radiometer generally worked properly. There were various times when cleaning or releveling the sensor affected the data, those bad data will be removed from the NetCDF files. There were also a few instances (~13 total from all towers) in which electrical shock, a loose connection, cable being chewed by cows, or yet undetermined reasons caused the removal of some data from the NetCDF files. In addition, a new net radiometer was installed on August 10th, 2006, at station 1, because the previous one was affected by electrical shock. This caused a slight descrepency in the data. If looked at closely, when comparing the net radiation values at station 1 to station 10 (the station with the closest proximity to station 1), one can see that the nighttime values at station 1 are slightly lower (more negative) than station 10 for the remainder of the project. This could be a result of a slightly different nighttime calibration of the newly installed net radiometer (click on this link for an example plot), but that is only a speculation.

Longwave radiation values calculated from the measured Rpile, Tcase, and Tdome values will be added to the NetCDF files (note that the Kipp&Zonen pyrgeometers do not have Tdome sensors).

The wind speed correction recommended by Campbell Scientific for the net radiometers will be implemented.


Soil Sensors (soil temperature, soil heat flux, soil moisture, soil thermal properties (TP01))

Soil Temperature:

The sensors generally worked properly. At station 5, a data glitch appearing from approximately June 27th to July 2nd caused one unrealistic point every three hours, those points will be removed from the NetCDF files.

At station 10, after the first significant rainfall event on July 26th overnight, the soil temperature signal began to gradually attenuate and level off for the rest of the project. Here is a link to a plot at station 10: Tsoil vs T. We presume this was caused by abundant grass growth having an insulating effect on the area around the soil temperature sensor. The grass didn't allow direct sunlight (shortwave radiation) to conduct effectively into the soil to warm it up during the day and didn't allow nighttime diffuse radiation (longwave radiation) to effectively exit the soil and cool it down at night in a "greenhouse" type effect. Please follow the links below for examples of before and after pictures illustrating the abundant grass growth after the first significant rainfall.


Soil Heat Flux

The sensors generally worked properly. There were a few instances (~7 total from 4 flux towers) when unknown reasons caused an unrealistic spike in the data, those will be removed from the NetCDF files. At station 5, a data glitch appearing from approximately June 27 - July 2 caused one unrealistic point every three hours, those points will be removed from the NetCDF files. At station 3 and 10, data glitches occuring at nighttime because of problems with the wireless Xbee dropping characters is in the process of being quality controlled. The approximate time when these data glitches occurred are as follows:
  • Station 4: August 20th, 16:25 - August 21st, 14:30
  • Station 5: July 19th, 18:15 - July 21st, 12:29

There were also a few instances (~13 total for all 10 towers) when rain infiltration, tower maintenance periods, or unknown reasons caused unrealistic values in the data, those too will be removed from the NetCDF files.


Soil Moisture:

The sensors generally worked properly. An extensive set of gravimetric samples were taken at each of the four sites to perform an in-situ calibration (to reduce any errors associated with associated soil properties).  It was necessary to make significant adjustments to obtain reasonable agreement with these samples:

  • NE (2): Qsoil_meas = 1.8 Qsoil_grav/Rho - 1%; Rho=1.30 g/cm^3
  • ENE (3): Qsoil_meas = 1.2 Qsoil_grav/Rho - 2%; Rho=1.24 g/cm^3
  • SSE (5): Qsoil_meas = 1.4 Qsoil_grav/Rho - 3%; Rho=1.67 g/cm^3
  • NNW (10): Qsoil_meas = 1.2 Qsoil_grav/Rho - 1%; Rho=1.50 g/cm^3

A plot of this comparison after these adjustments is shown below.  However, these adjustments have not yet been applied to the data.

There were a few instances (~3 total from 4 flux towers) when unknown reasons caused an unrealistic spike in the data, those will be removed from the NetCDF files. At station 5, a data glitch appearing from approximately June 27th to July 2nd, caused one unrealistic point every three hours, those points will be removed from the NetCDF files.


Soil Thermal Properties (TP01):

The sensors appeared to work properly, however there was a logbook comment after tear-down: "The TP01 probes, at site 2, 5, and 10, had a fold in the thin film transducer. I assume during installation that the top of the film caught an edge and started the fold. This is something nobody would have noticed."  Since the derivation of heat capacity values depend on the probe being straight, the data from these sites should be used with caution.  Values from site 3 should be fine.

Values for the soil heat capacity, thermal diffusivity, and thermal conductivity derived from the raw TP01 readings will be added to the NetCDF files.


GOES Communications

Each station transmits a message every 5 minutes, on channel 95 of GOES East, in its own time slot, specified by an offset in seconds after the start of each 5 minute interval. Basic stations send 46 bytes of data, which takes about 5 seconds at the GOES RF baud rate of 100 bps. Flux stations send 162 bytes (~16 seconds), and the Flux/CO2 station (SSE) sends 182 bytes (~18 seconds).
Station GOES Id Transmission Offset (sec) Transmission Length (sec)
1 3640C550 165 5
2 3640D626 30 16
3 364102B4 60 16
4 3641372E 90 5
5 36417424 120 18
6 3641A24C 150 5
7 3641B13A 180 5
8 3641D4DC 210 5
9 3641E146 240 5
10 3641F230 270 16

Data Tables and Plots

Click on the link to view tables and plots.

The tables display mean, deviation, number of missing values, and number of clipped values for each variable, from 0:00-6:00, 6-12:00, 12-18:00 and 18-24:00 each day.

The 5 minute statistics for most variables are shown on time-series plots.


Data Download

The NCAR data are available for download in the following forms:We will also have a Table of Variables available in these files.




Field Logbook

A computer-readable field logbook of comments by NCAR and other personnel is available in raw text format.