The level of reduction is represented in the various possible locations in the data tree where information is stored.

E.g. for a monochromatic beam it is sufficient to supply the wavelength information (value, resolution and unit) only once. The right place is the experiment/measurement section. But for a energy-dispersive set-up the wavelength might be given for each measurement point in and extra column.

Below several examples of hierarchical structures are given for various measurement schemes and levels of reduction. Shown are only the entries relevant for this discussion.

monochromatic, angle-dispersive scheme

The wavelength and its resolution are given only once because they are constant. But the angle points to the column where the individual values are given. These allow for footprint correction.

experiment:
    measurement:
        scheme:     angle-dispersive
        wavelength: {value: 4.0, resolution: {type: constant, sigma: 0.1}, unit: angstrom}
        angle:      {column: 5}
reduction:
    input files:
        - file: data1
        - file: data2
columns:
    - {column: 1, variable: Qz, unit: 1/angstrom}
    - {column: 2, variable: R}
    - {column: 3, variable: sR}
    - {column: 4, variable: sQz, unit: 1/angstrom}
    - {column: 5, variable: angle, unit: deg}

energy-dispersive scheme, one angle

The wavelength column allows for absorption correction.

experiment:
    measurement:
        scheme: energy-dispersive
        wavelength: {column: 5}
        angle: {value: 1.2, resolution: {type: const, sigma: 0.02}, unit: deg}
reduction:
    input files:
        - file: data1
        - file: data2
columns:
    - {column: 1, variable: Qz, unit: 1/angstrom}
    - {column: 2, variable: R}
    - {column: 3, variable: sR}
    - {column: 4, variable: sQz, unit: 1/angstrom}
    - {column: 5, variable: wavelength, unit: angstrom}

energy-dispersive scheme, various angular settings

The wavelength information applies for all input files, thus it is in the experiment/measurement section. The angle here describes the individual raw data files. Both entries are information for the user, since the level of reduction assumed here (merging of the data form both raw files) does not allow for later absorption or footprint corrections.

experiment:
    measurement:
        scheme: energy-dispersive
        wavelength: {min: 3.0, max: 12.0, resolution: {type: const, sigma: 0.1}, unit: angstrom}
reduction:
    input files:
        - file: data1
          angle: {value: 1.2, spread: 0.02, unit: deg}
        - file: data2
          angle: {value: 2.4, spread: 0.02, unit: deg}
columns:
    - {column: 1, variable: Qz, unit: 1/angstrom}
    - {column: 2, variable: R}
    - {column: 3, variable: sR}
    - {column: 4, variable: sQz, unit: 1/angstrom}

angle- and energy-dispersive, various angular settings, compact

Data are re-binned and histogrammed to a (given) Qz grid. Resolution information is partially lost.

The universal wavelength and the raw file specific angle are descriptive. They can hardly be used for data analysis.

experiment:
    measurement:
        scheme: angle- and energy-dispersive
        wavelength: {min: 3.0, max: 12.0, spread: 0.1, unit: angstrom}
reduction:
    input files:
        - file: data1
          angle: (min: 0.6, max: 2.0, unit: deg}
        - file: data2
          angle: (min: 2.4, Max: 3.8, unit: deg}
columns:
    - {column: 1, variable: Qz, unit: 1/angstrom}
    - {column: 2, variable: R}
    - {column: 3, variable: sR)}
    - {column: 4, variable: sQz, unit: 1/angstrom}

angle- and energy-dispersive, various angular settings, detailed

Data might be re-binned, but are not histogrammed. I.e. there might be several entries with (about) the same Qz, but with different wavelength and angle and thus sQz.

The analysis software might use either sQz or the (wavelength, angle) tuple, where the resolution information for both is given in the universal part. Besides this, the universal wavelength and the raw file specific angle are descriptive and not meant for data analysis.

experiment:
    measurement:
        scheme: angle- and energy-dispersive
        wavelength: 
            min: 3.0
            max: 12.0
            resolution: {type: prop, sigma: 0.02}
            unit: angstrom
            column: 5
        angle: 
            resolution: {type: const, sigma: 0.01}
            unit: deg
            column: 6
reduction:
    input files:
        - file: data1
          angle: (min: 0.6, max: 2.0, unit: deg}
        - file: data2
          angle: (min: 2.4, max: 3.8, unit: deg}
columns:
    - {column: 1, variable: Qz, unit: 1/angstrom}
    - {column: 2, variable: R}
    - {column: 3, variable: sR}
    - {column: 4, variable: sQz, unit: 1/angstrom}
    - {column: 5, variable: wavelength, unit: angstrom}
    - {column: 6, variable: angle, unit: deg}

tasks

  • What are the consequences for the reduction software?
  • What is the consequence for the analysis software when the information might be found in various places in the file?

decision