OVRO-LWA Data Products

Introduction

THIS PAGE IS CURRENTLY UNDER CONSTRUCTION.

OVRO-LWA is an all-sky radio imager by design and hence in principle, can observe the Sun as long as it is above the horizon. However, the array is located in a valley surrounded by mountains, and hence the Sun is not visible from the array when its line-of-sight from the array passes through the mountains. Additionally at very low elevations, the performance of the array degrades significantly and the data gets increasingly affected by terrestrial radio emission. Keeping this in mind, the Sun is observed with the OVRO-LWA when the solar elevation is greater than approximately about 15 degrees. In summer this translates to about 7-14 hours of observations depending on the season. The highest time-frequency resolution at which data can be obtained with the OVRO-LWA in a regular manner is 1ms and 24 kHz respectively. However depending on the observation mode as well as due to data volume limitations, the actual available time-frequency resolution can vary. Figure 1 summarizes the different levels of data we produce. The later sections will give a more detailed description and usage examples.

Level 0 - Raw data from the instrument

OVRO-LWA, in general, operates multiple observing modes simultaneously. This is achieved by passing the raw data stream from the 352 antennas through multiple data handling processes, with each process handling an observation mode. The middle panel of Figure 1 (indicated as Level 0) shows the key parameters relevant for the solar data recorded by the different data streams. The imaging data and dynamic spectrum data refers to the visibility data and coherently beamformed data respectively.

Level 1.0 - Images and spectrogram data

Due to the high data volume, no level 0 visibility data are stored for long-term. The standard data processing pipeline processes the level 0 data and converts them into images. After the processing, the level 0 visibility data are deleted. While the level 0 dynamic spectrum data are stored, again due to the high data volumes, these data will only be provided to the community upon request. In Figure 1, in the right panel, we have indicated the level 1 data products which will be provided to the community by the OVRO-LWA team in a regular manner.

The images are stored and provided in HDF5 format. The OVRO-LWA team also provides software to convert the HDF5 files to more readily usable FITS file format. The produced FITS files, apart from the multi-frequency data, also contains a table containing the frequencies corresponding to the multi-frequency images and the instrumental resolution at each frequency. It also contains some other parameters, which are necessary to convert the images to level 1.5, which might be useful for some scientific purposes. These images are in heliocentric coordinates and typically have a dynamic range of 300 and higher. Dynamic range refers to the ratio of the image maximum and the uncertainties, which is typically quantified by the image rms. The fits images can be directly loaded into Python or SSWIDL using standard techniques. The dynamic spectrum data is provided in standard FITS format.

Below we describe each of these level 1 data products.

• Imaging data products:
  • Frequency integrated images: 12 images, with center frequencies of 34, 39, 43, 48, 52, 57, 62, 66, 71, 75, 80 and 84 MHz, are provided. The images are produced by integrating 10s and approximately 5 MHz of data. The cadence of the images vary from 10s to sometimes a minute.

• • High frequency resolution images: 144 images, with frequencies ranging from 32 - 88 MHz, with frequency integration of 416 kHz. The imaging cadence and time integration is the same as that of the frequency integrated images.

• Dynamic spectrum data product: The dynamic spectrum data are provided as standard FITS tables containing the frequency list, list of times, and the Stokes I flux density in SFU.

Level 1.5 - Images

Low radio frequency waves, during propagation, get significantly affected by ionospheric refraction. Ionospheric refraction often results in source shifts only, without any change of the source morphology and flux density. In this case, it can be shown that the source shift is inversely proportional to the square of the observation frequency. Here this dependence is used to try to correct for refraction. The proportionality constant is directly dependent on the gradient of the ionospheric total electron count along the line-of-sight towards the source. When the sun is quiet, the imaging dynamic range is sufficiently high to easily see the quiet sun disc. Hence the ionospheric source shift can be determined by comparing the observed center of the solar disc with the optical location of the Sun. The source shifts at multiple frequencies can be fitted to obtain the ionospheric parameters, and then can be applied to the images at other frequencies, where the solar disc is not well observed due to dynamic range limitations.