Table of Contents
Last updated: 4 February, 2016
All Herschel data comes from the HSA in the form of an ObservationContext, which is comprised of various layers of science data, auxiliary data, and other information about your observation in an onion skin type of structure. In this chapter we describe what you will find in an HIFI ObservationContext. First we give a short introduction to the structure of Herschel HIFI data storage, starting with the smallest data structure, the dataframe, and work our way up to the outer skin of the Observation Context.
The Herschel spacecraft stored data onboard (up to two days' worth) until it was transmitted to Earth. Science data, such as a WBS spectrometer readout, came naturally in sets, or Frames. Data frames were packetised for transmission from the Herschel Space Observatory to Earth. Along with House Keeping (HK) data, they were downlinked to the tracking station and thence to the Mission Operation Centre (MOC) at ESOC in Darmstadt, or to the latter directly. The data packets then flown from the MOC to the Herschel Science Centre (HSC) at ESA's European Space Astronomy Centre (ESAC) in Madrid. The HIFI ICC copied the data from HSC, as well.
At ESAC, the data packets were 'ingested' into a database and the science data frames were reconstituted.
The combination of HK and science data created a SpectrumDataset, or spectrum in more common language.
A Herschel Data Product consists of metadata keywords, tables of SpectrumDatasets, and the history of the processing that generated the product. There are various product types (Observation, Calibration, Auxiliary, Quality Control, User Generated).
The types of Observation Data Product for HIFI is called a HifiTimelineProduct, or HTP. This is a time-ordered series of SpectrumDatasets comprising all the integrations in the observation.
For more information about Herschel products, see the Product Definitions Document .
A Context is a subclass of Product, a structure containing references to Products and necessary metadata. A Context can contain Contexts, giving rise to Context 'trees.' Types:
ListContexts (for grouping products into sequences or lists, hardly used)
MapContexts (for grouping products into key value dictionaries)
A MapContext instance serves as the organisational product unit for the Herschel Data Processing system. It contains the following contexts:
Level-0, Level-0.5, Level-1, Level-2, Level-2.5, & Level-3 (optional) Contexts
Trend Analysis Context
optional Telemetry Context: not by default, only when the HSC deems it necessary because of a serious problem in the processing to Level-0 data.
The uses of these Contexts will be described in Chapter 5.
Note that the descriptive modifiers "Product" and "Context" are often dropped conversationally.