Visualizing SED Data in Iris

When SED data is read into Iris, it displays in the Iris Visualizer, where you can interact with the data plot in various ways. The options available for customizing the data display are described in this thread.

When multiple segments and/or photometric points have been read into the Iris session and coplotted in the main display, it is important to note that the spectral data are not combined, coadded, or spliced in memory in any way; the raw data from the multiple input spectrograms/points is completely preserved in the resulting combination. This means that two separate SED data segments loaded into Iris for a particular source - e.g., observed with two different observatories and spanning separate (or overlapping) spectral ranges - may be offset from one another, as Iris does not scale, trim, and join the separate segments to form a seamless SED for the source. This is illustrated in the image below.

Clicking the "Metadata" button in the upper-right corner of the Iris display opens a window with tabs containing different levels of data and metadata in the SED.

Point Metadata tab

When available, the Metadata tab includes many useful pieces of information about the data points currently displayed. When a SED is imported from NED, as in the example considered in this thread, the metadata includes such things as the bibliographic reference code for each data point, the spectral range covered by instrument used to obtain data point, data point uncertainty and flux values as they are published, data point significance values, among other properties. The full list of metadata information available for each data point will depend on the specific data sources.

The metadata table may be sorted by clicking the header of the column by which you would like to sort - once for ascending order, twice for descending order, and three times to restore the default sorting - and rearranged by clicking and dragging columns left or right. A nested sort may be achieved by first selecting the column which will set the master sort, clicking and holding the Control key, and then selecting the column by which to sort the groups of the master sort.

Segment Metadata tab

Metadata that is common to all data points in a SED segment is displayed under this tab, in a similar table as the one that appears under the Point Metadata tab.

Data tab

The X and Y coordinate values of each SED data point in the Iris display are contained in the "Spectral Axis" and "Flux Axis" columns of the Data tab in the Metadata window, and reflect the values as they were imported into Iris from NED or uploaded from a file on your hard disk - i.e., not what is currently plotted in the Iris display (in the event that you changed the units of the data plot within the display). For example, if the data were uploaded in Jansky flux units versus frequency in Hertz, but then you change the display units to ergs/s/cm2/Angstrom versus Angstrom, the data point values returned will be in Jansky flux units and Hertz.

Data points can be masked, or 'grayed out' from the Iris plot display with a number of tools. The most basic works by simply selecting the rows corresponding to these points in any one of the Point Metadata, Segment Metadata, or Data tabs of the Metadata window, and then click "View un-selected" at the bottom of that window. In the display, the points selected in the Metadata window will appear fainter than the unselected points.

The "View selected" button performs the reverse operation, and the "Restore" button restores the plot display to its original state. The "Un-select" button removes the selection state from all rows in the table, without affecting the plot display.

Note that these data masking operations relate to the display only. Internally, masked data is kept intact and will be used in data analysis and modeling performed by Iris. To really get rid of unwanted data, please refer to section 'Extracting a Filtered SED' below.

Data point selection can also be performed based on the plot itself, instead of the table. One way to do that is by clicking on a specific data point on the display. The corresponding row in both the Point Metadata and Data tables will be selected. This selection is non-destructive, meaning that by successively clicking on points on the display, one can add the rows corresponding to each one, to the set of selected rows.

Another way is by zooming over the region that contains the points one wants to select, and then clicking the "Select from plot" button on the Metadata window. The table rows corresponding to the data points that show up inside the plot viewport will be added to the selected set of rows already on the table.

Selection by Boolean operations

The Point Metadata table can be hierarchically sorted by clicking on specific column headers. Clicking once sorts the rows in ascending order of that column; next click sorts it in descending order, and the next click places the rows back on their original ordering. By holding the Ctrl key pressed when clicking on a column header, the sorting state of previously sorted columns is kept unchanged, thus enabling hierarchical sorting.

Using that mechanism, rows can be re-ordered together according to relatively complex selection criteria against column content. This facilitates the lumping together of the desired rows, that can in turn be selected and operated upon with the tools described in the preceding paragraph.

Note that columns can be re-positioned at will by dragging their headers horizontally.

Note also that string-valued columns are treated in lexical order. Integer and floating-point-valued columns are treated in numerical order.

In order to enable even more complex selection criteria though, table rows can be selected based on the result of an arbitrary Boolean expression computed on selected column contents. This expression is entered in the 'Type boolean expression:' text field, and by hitting Return, or clicking on the 'Select points' button, the expression is computed for every row in the table, and the row is selected if the Boolean result is True.

Iris uses a Python interpreter to parse and execute the expression, thus the syntax for this expression is plain Python. All Python built-in and string functions are supported. See the Python documentation here: http://docs.python.org/2/library/functions.html http://docs.python.org/2/library/string.html#string-function

In the expression, columns are referred by their name prefix. Notice that each column name starts with a lower-case letter followed by a colon (':'). This is the prefix one should use in the expression to refer to the column.

An example of a valid expression that can be applied to the NED-supplied SED for 3C 066A could be:

l: > 1.0 and c:.rstrip().endswith('reported')

In this SED, column l: contains the DataFluxPublishedValue metadata property, a floating point value, and column c: contains the DataSignificance metadata property, a string.

Of course, in this very simple example the same selection could be accomplished by just sorting the table by column c: first, and then holding the Ctrl key and sorting by column l:. The desired rows would be easy to spot as a single group or contiguous rows, and could be selected with a simple mouse gesture.

The "Extract" function in the Metadata window allows you to go one step further than simply masking unwanted data points; it allows you to extract a whole new SED consisting of only the selected data points.

Making the desired row selections in the Metadata window and then clicking "Extract" will open a new SED in the SED Builder window named "FilterSED" - an ID which you can change - which will display in the Iris Visualizer.

The drop selector at the right bottom of the Point Metadata and Data tabs allows one to pick a specific column and have Iris paint each data point on the display with a color that linearly maps into the data range of the chosen column.

Floating point columns are mapped continuously into color space. Integer and string valued columns are mapped in such a way that the number of different colors used in the plot is the same as the number of unique values in the column. String-valued columns are mapped lexically into color space.

The color space is defined here

By right-clicking (Ctrl-click on MacOS) on a specific data point on the Iris display plot, a window containing information for that point is brought to screen. The window contains tabs that display similar information as described above for the Point Metadata, Segment Metadata, and Data tabs for the entire SED, but relative to the selected point only.

This window has an extra tab named 'Aperture Correction' that enables one to apply a simple multiplicative flux correction to the data point.

By entering either the desired flux, or the ratio of the desired flux to the original flux, and clicking the appropriate button, the data point flux will be modified. Iris keeps the original flux so successive applications of this tool will not cause the original flux value to be lost. In that way, we can always assign a precise meaning to the ratio value.

The correction can be applied as well to the entire segment to which the point belongs to. Just select the Segments radio button and enter the desired flux ratio.

One can also apply a flux correction by directly dragging the desired point on the Iris display plot, holding the Shift key down while dragging. Note that this feature is always enabled, even when no Aperture Correction window is on screen.