In the following we describe the contents and formats of the data in the compressed file http://sts.bwk.tue.nl/drivingrain/data_199712_200003.tgz. The data form an important result of a research on driving rain at the TUE. Please read the copyright notice below before use of the data.
The description of the measurement set-up and the analyses of the data are reported in Driving rain on building envelopes by Fabien J.R. van Mook (Technische Universiteit Eindhoven, 2002). This thesis is available in PDF-format and on-line at http://sts.bwk.tue.nl/drivingrain/.
The directory structure is as follows:
One is referred to chapter 3 of Driving rain on building envelopes for an extensive description of the measurement set-up, instrumentation, data acquisition systems and data processing methods. The following list summarises the measurement devices and their operational periods more exactly than in chapter 3. One is referred to the mentioned chapter for the location of the measurement positions P1 to P7.
For temperary malfunctions of a device, one is referred to the logbook: /info/logbook.txt. It lists also the moments when a device is installed, checked or cleaned, when the whole system malfunctioned, when interesting events (weather) occured and so fourth. Although the measurement data provided hereby start at 1997-12-01, the logbook mentions concisely some of the preparatory work (installation of some of the devices; first tests of the data acquisition system).
The raw data of the disdrometer will be treated differently from the data on wind, horizontal rain and driving rain. Note that we use the notations <year>, <month>, <day>, <hour> etc. to indicate the time format. The time indications are always UTC+1h.
The files of the raw data on wind, horizontal rain and driving rain are «thinned-out» versions of the original data files. The original data files contain every individual reading, i.e. from every polling to a device by the data acquisition systems. Moreover, the format of the original data files are quite redundant. Examples of original data files are /info/99120100.phy (the anemometer at P1) and /info/99120106.001 (the other devices). The original data files are thinned out by discarding repetitions, other redundancies and irrelevant information. It was also necessary to reduce the wind velocity measurements at P4, which were sampled every second, by averaging the velocity components to 1-min values.
The raw data files reside in the directory /data. It is subdivided in subdirectories for every month, e.g. /data/9808 for August 1998. Every subdirectory contains 15 (plain ascii) files:
<day> <hour> <minute> <second> <sampletime> <U1> <U2> <U3>with <U1>, <U2> and <U3> = 1-min averages of the wind velocity components in cm/s according to the axis system of the Solent Research anemometer (see section 3.2.1 in Driving rain on building envelopes), and <sampletime> = the sample interval in s.
<day> <hour> <minute> <U1> <U2> <U3> <n>with <U1>, <U2> and <U3> = 1-min averages of the wind velocity components in m/s according to the axis system of the Solent Windmaster anemometer (see section 3.2.1 in Driving rain on building envelopes), and with <n> = the number of succesful readings. Note that (1) the original sample rate was 1 per second, (2) a line was not written if there were no readings during the 1-min clock period, and (3) the distance of the anemometer to the building façade is not indicated in the file (one is referred to the section Measurement devices).
<day> <hour> <minute> <second> <sampletime> <tips>or:
<day> <hour> <minute> <second> 8889 0or:
<day> <hour> <minute> <second> 8890 0or:
<day> <hour> <minute> <second> 8891 <bucketvolume>with <tips> = the number of tippings, and <bucketvolume> = the volume of the tipping bucket in ml. A line of the first format was only written when the number of tips is more than zero. A line of the second and third format was written when the device was put on or off, respectively. A line of the fourth format was written at the beginning of the file, and when the bucket volume changed.
<day> <hour> <minute> <second> <sampletime> 1or, when the rain indicator was put on:
<day> <hour> <minute> <second> 8889 0or, when the rain indicator was put off:
<day> <hour> <minute> <second> 8890 0
<day> <hour> <minute> <second> <sampletime> <weight>or:
<day> <hour> <minute> <second> 8889 0or:
<day> <hour> <minute> <second> 8890 0or:
<day> <hour> <minute> <second> 8891 <catchmentarea>with <weight> = the weight in grams of the reservoir on the balance (i.e. the reservoir itself + the collected driving rain water), and <catchmentarea> = the catchment area in m2. A line of the first format was written at the beginning of the file, and when the weight changed. A line of the second and third format was written when the device was put on or off, respectively. A line of the fourth format was written at the beginning of the file, and when the catchment area changed.
<day> <hour> <minute> <second> <sampletime> <tips>or:
<day> <hour> <minute> <second> 8889 0or:
<day> <hour> <minute> <second> 8890 0or:
<day> <hour> <minute> <second> 8891 <bucketvolume>with <tips> = the number of tippings, and <bucketvolume> = the volume of the tipping bucket in ml. A line of the first format was only written when the number of tips was more than zero. A line of the second and third format was written when the device was put on or off, respectively. A line of the fourth format was written at the beginning of the file, and when the bucket volume changed.
<day> <hour> <minute> <second> <sampletime> <mvolts>or:
<day> <hour> <minute> <second> 8889 0or:
<day> <hour> <minute> <second> 8890 0or:
<day> <hour> <minute> <second> 8891 <calibrationconstant>with <mvolts> = the voltage from the strain gauge in mV (proportional to the weight of the collector + collected driving rain water), and <calibrationconstant> = the calibration contant in mV/g for the convertion of <mvolts> into the weight of the collector. A line of the first format was written for every reading. A line of the second and third format was written when the device was put on or off, respectively. A line of the fourth format was written at the beginning of the file, and when the calibration contant changed.
<day> <hour> <minute> <n10> <n104> <n105> <n204> <n205> <n206> <n40> <n41> <n60> <n61>with <n10>, <n104>, <n105> etc. = the number of readings per 5-min clock period of device 10, 104, 105 etc. (i.e. the anemometer at P1, the TUE-I gauge at P4, etc. -- the figures correspond to the figures after the underscore in the file names). A line was not written if there were no readings during the particular clock period at all.
<year>-<month>-<day> <hour>:<minute>:<second> <devicecode> ON <comment>or:
<year>-<month>-<day> <hour>:<minute>:<second> <devicecode> OFF <comment>with <devicecode> = a code for the device (explained in /info/logbook.txt).
With the files <year><month>_reg.dat one can evaluate whether a device worked properly during a 5-min clock period. These files are necessary because for some of the devices (e.g. the tipping-bucket rain gauges) the zero readings or unchanged readings are omitted in the raw data files provided hereby, while they appeared in the original raw data files.
The raw disdrometer data are stored in the directory /disdro. The disdrometer was operational from 1999-10-01 09:30 till 2000-01-11 15:30. The data between 1999-12-11 14:40 and 1999-12-20 13:00 were unfortunately lost during downloading from the data processing unit of the disdrometer. The data from 2000-01-07 00:38 till 2000-01-11 11:45 were thought to be lost (and never used in Driving rain on building envelopes), but could still be saved after the dismounting of the disdrometer off the roof of the Auditorium.
The raw disdromater data files in the directory /disdro were left in their original format as given by the disdrometer. An explanation of the format is given in section 4 «Description of data telegram» in /info/disdro.txt. Sections 5 and 6 of the manual of the disdrometer are also important.
Directories /datac5 and /datac10 contain the files of the processed data with a clock period of 5 and 10 min, respectively. They result from processing the original raw data according to the methods mentioned in sections 3.4 and 5.1 of Driving rain on building envelopes. The processed data files were used for the analyses in Driving rain on building envelopes.
Directory /datadis contains the files of the processed data with a clock period of 5 min and for the period 1999-10-01 to 2000-01-31. They are obtained in the same way as those in directory /datac5, but contain the processed disdrometer data in addition.
Important to note is the notion «availability» of a particular clock period. In secion 5.1 it is said that a clock period is said to be «available» if every device, responsible for one or two of the desired quantities, is working properly. A device which is simply not installed (operational), does not cause a clock period to be unavailable. In the section Measurement devices the operational periods of the devices are listed. An out-of-order due to maintenance, power failure or other malfunction when a device was operational, causes a clock period to be unavailable. A clock period is also unavailable when a device produced less than 90% of the readings which it should have produced during a clock period because of its sample time.
The files of the processed data are in plain ascii. Their filenames and contents are listed below. A line starting with a percentage-sign (%) contains a comment. Discarding these comment lines, one obtains files in which every nth line corresponds to each other, namely to the same clock period. A file contains the processed data for one or more quantities of the whole 24-month measurement period (1997-12-01 to 1999-11-30).
A value of 8888 for a particular quantity at a particular clock period means that this particular measurement is not available.
<year> <day> <month> <hour> <minute> <second>
<Uh> <F> <Uy> <Uz>with <Uh> = horizontal wind speed in m/s, <F> = horizontal wind direction in degrees clockwise from north, <Uy> = wind velocity component Uy in m/s, and <Uz> = vertical wind velocity component in m/s.
<distance> <Ux> <Uy> <Uz>
One is allowed to use the data only if the university and the Building Physics group are acknowledged by mentioning the copyright holder (Technische Universiteit Eindhoven, Fysische Aspekten van de Gebouwde Omgeving), the website (http://sts.bwk.tue.nl/drivingrain/) and the date of publication (2002), or by referring to Driving rain on building envelopes by Fabien J.R. van Mook (Technische Universiteit Eindhoven, 2002).
The data is provided by the copyright holder «as is», without any express or implied warranties of any kind. In no event shall the copyright holder be held liable for any damages connected with the use of this program.
A short message to prof.dr.ir. Martin H. de Wit <M.H.d.Wit@bwk.tue.nl> is appreciated when you use the data for simulations, comparisons, driving rain models, etc.
For questions and remarks one can contact him too. His address is:
Martin H. de Wit
Building Physics group (FAGO)
Technische Universiteit Eindhoven
Postbus 513
NL-5600 MB Eindhoven
the Netherlands
Telephone: +31-40-2472557
Fax: +31-40-2438595
E-mail: M.H.d.Wit@bwk.tue.nl
© December 2002 Technische Universiteit Eindhoven/ Fysische Aspekten van de Gebouwde Omgeving/ Fabien van Mook