3 GHz reflectivity data in Hierarchical Data Format (HDF)

This page describes the scanning data recorded by the 3 GHz Chilbolton Advanced Meteorological Radar (CAMRa) during the CLARE98 flights.

The program to convert the standard Chilbolton radar data format to HDF was written by Luke Robertson at RAL. As a fully working version of this has only been available since January 1999, the IDL code so far written to extract scans from HDF files is in rather a primitive stage. It is also possible that there could be problems with the data itself. If you have any problems extracting scans or suspect there is a problem with the data, please contact Robin Hogan. Note that the data in these files are not calibrated - see the section on Calibration below. The calibration figure has also changed (as of May 11, 1999).

Layout of the data

Each Chilbolton file consists of a number of rasters, each of which contains one or more scans. A scan is one of the following:

RHI (Range-Height Indicator) - a scan in elevation with a constant scan-rate
PPI (Plan-Position Indicator) - a scan in azimuth with a constant scan-rate
CSP (Custom Slant Plane) - a scan which can vary in both azimuth and elevation
FIX - a dwell at fixed azimuth and elevation

When there are several scans in one raster, they are each of the same type and cover the same range of angles (during CLARE98 at least).
During CLARE98 the `Custom Slant Plane' was only used for vertical scans along the azimuth of the aircraft, with a varying scan-rate designed to achieve coincidence between the radar and the C-130 aircraft. Note: on some occasions the aircraft is clearly visible to the radar as a `streak' at a particular altitude. This does not necessarily indicate the exact altitude of the aircraft, as it is a large enough target to be detected in the side-lobes of the radar.

Click here to download a postscript figure showing schematically how the data is organised in the HDF files.

Downloading the data

The data can be found as gzipped HDF files in the subdirectory hdf. Click on the following links to download a file. The gzipped sizes are shown alongside. The files approximately double in size when unzipped.

camra_981007.hdf.gz (Chilbolton file 6133) 31 MBytes
camra_981013.hdf.gz (Chilbolton file 6139) 46 MBytes
camra_981014.hdf.gz (Chilbolton file 6140) 36 MBytes
camra_981016.hdf.gz (Chilbolton file 6143) 26 MBytes
camra_981020.hdf.gz (Chilbolton file 6146) 77 MBytes
camra_981021a.hdf.gz (Chilbolton file 6148) 37 MBytes
camra_981021b.hdf.gz (Chilbolton file 6149) 12 MBytes
camra_981022.hdf.gz (Chilbolton file 6151) 27 MBytes

These files contain Reflectivity Z, Differential Reflectivity Z_DR and Linear Depolarisation Ratio LDR.

Also in the hdf subdirectory are corresponding ASCII files that consist simply of tables of numbers summarising each raster in the file. Note: these are not the same as the standard Chilbolton summary files (*.sum), which for the CLARE98 flights contained some misleading/incorrect/irrelevant entries. Click on the following links to download.

camra_981007.ascii
camra_981013.ascii
camra_981014.ascii
camra_981016.ascii
camra_981020.ascii
camra_981021a.ascii
camra_981021b.ascii
camra_981022.ascii

Each row in the file describes a raster. The columns correspond to the following:

raster number
time, UTC (hh:mm:ss)
date (dd:mm:yyyy)
scan type (RHI/PPI/CSP/FIX)
range gate spacing (75/150/300 metres)
minimum angle in raster (minimum ELEVATION for RHI/CSP/FIX, minimum AZIMUTH for PPIs)
maximum angle in raster (maximum ELEVATION for RHI/CSP/FIX, maximum AZIMUTH for PPIs)
`initial' angle in raster (AZIMUTH for RHI/CSP/FIX, ELEVATION for PPIs)
number of scans (Note: often scan sequences are aborted half way through, in which case the true number of scans is less than that reported in this file)

Calibration and other considerations

NOTE: This following calibration figure for Z has changed from the original figure of 4 dB in light of the most recent calibration; click here for the most recent calibration document.
Z: Add 5 dB to the raw radar reflectivity to obtain the correct values
Z_DR: Add 1 dB to the raw differential reflectivity to obtain the correct values

Note that ground clutter is quite significant in the nearest 10 km, and can make the data here unusable.
The contribution from thermal/receiver noise noise has not been removed from the data

Reading in the data with IDL

This directory contains two sample IDL programs to read in scans. They should be regarded as an example of how to read in the data, and no doubt users will wish to write their own programs to do this. They rely on the HDF procedures intrinsic to IDL, and are not guaranteed to work with versions earlier than 5.2 (although I seem to be able to read them with version 5.0 no problem).

The first program is hdf_read.pro. To load in a scan:

.run hdf_read
data=samplecode()

You are then presented with a window to select the HDF file you wish to read from. The program will then read how many rasters there are in the file and ask you to select one. Note: at this stage the list of raster numbers starts from 1 and increases contiguously, so does not account for missing rasters. Therefore the raster numbers may be different from those in the `.ascii' files. You are then asked to select which scan you are require. Finally the program writes `data', a pointer to a structure containing the individual parameters. The structure consists of

pheader - pointer to the standard Chilbolton header (another structure), although not many of the header fields are being set at this stage
pAz - Pointer to array containing azimuth positions for this scan in degrees clockwise from DUE EAST
pEl - Pointer to array containing elevation positions for this scan
pGateRange - Array containing gate ranges
CountParams - Number of radar parameters stored
pData - Array of pointers to the radar parameters
DataTypes - Describes what data is stored

The second program, get_variables.pro is a simple demonstration of how to uses the pointer data to create workspace variables of each radar parameter. Have a look at get_variables.pro to see what parameters are extracted, and their units. Type:

.run get_variables

Then for an RHI you could roughly plot the scan with

contour, Zh, r, z, /fill

And for a PPI you could use

contour, Zh, x, y, /fill