BackThe incorporation of Stellarium charts into the Parades page you've just come from was not as easy as might be imagined. It was not, for example, merely a case of setting up a particular time, date & place in Stellarium and asking it to output a labelled chart of the sky corresponding to those details. This is mainly because Stellarium is highly configurable, which is at the same time a blessing and a curse. A blessing because anything you want to configure probably can be, but a curse because it's not always easy to find out how to! Consequently, until one is well-versed in how Stellarium works it's sometimes easier to do things "by hand" rather than delve into the menus & user guide to find out how to persuade Stellarium to do it for you. The following is therefore not to be taken as "how to do it" but rather "how it was done on this occasion".
The first issue to confront was setting up the display parameters to generate a view which encompassed the entire sky from east to west, in order to show the Parades as a panorama, using a geometry which best imitated how the sky would be seen by eye - the best projection for this turned out to be Mercator. The display of such things as grids, the Ecliptic and the Celestial Equator also had to be turned off, again to ensure that the chart showed the sky as it would be seen "by eye".
Moving on to an issue connected with time, date & place, Stellarium tries to show the sky as it would actually be seen rather than just placing the celestial objects against a uniform black background. While this is undoubtedly helpful to an observer, to show whether a given object would actually be visible, it does mean that in charts set for times close to sunrise or sunset Stellarium will show a bright sky to the east or west respectively. This could result in planets "disappearing into the background" if they happened to be in that part of the sky.
Adding labels to the objects in view was the next problem. Although Stellarium can indeed label celestial objects, and the labels can be formatted by the user to some extent, restrictions on what can be changed mean that the labels for stars generally tend to be more prominent than for planets, which is the reverse of what would be preferable for a chart of a Parade - which is a parade of planets, after all. Also, as the placement of the labels seemingly cannot be changed, those for adjacent objects can sometimes overlap, with consequent reduction of legibility.
And finally, once everything had been set up as required how was the final chart to be output for use on the webpage?
Before discussing how were all these issues were overcome, I must firstly confess that they were not overcome by me personally as all the charts were produced by my astro-collaborator Roger, whose panorama is shown on the previous page. We did confer as to what might be desirable but it was he who did the "spadework".
A straightforward but effective way of generating output images was by the old technique of the screenshot - set up the view as required and use the facilities provided by the operating system to generate an image of what is seen on the screen. The required part of the image can then be prepared for use by judicious cropping. The only practical solution to the labelling issue seemed to be to format the labels by hand once the screenshot had been output and cropped. Using a graphics program (PaintShop Pro 7 in fact), Roger selected each part of each letter of the text of each label of a particular type of object using the "Magic Wand" tool together with the "add selection" function, and altered the colour & brightness of the selected portions to suit. Overlapping was solved by further magic - of the "cut-and-paste" type this time! The original labels were removed by using the clone brush to paint the nearby blank sky over them, then similar labels were copied from other charts and pasted in so they no longer overlap. Finally, the appearance of each label was adjusted slightly so they properly blended in with their new surroundings.
Which just leaves the "bright skies" problem. Fortunately, the only Parade for which this was a severe issue was that of February 2034, when Venus and Jupiter will be extremely low down in the sky at sunset and so the chart for the optimum day would be too bright in the west to show either of them. However, Parades tend to persist for a number of days and so although a chart might not show the optimum observing situation it can still show one which is representative. Roger and I thus experimented with possible options and found that if we went two days back from the optimum day and about 25 minutes later in time both planets became more easily visible against a much less bright sky. A reasonable solution, we felt.
That was not quite the end of the story though because to make changing the colour & brightness of the labels easier Roger had worked with a "full size" version of the charts i.e. the size produced by the screenshot. The final charts for display on the webpage would have to be considerably smaller and so size reduction by re-sampling was required. I then found that the judicious application of a bit of sharpening considerably improved the legibility of the now-smaller labels. And that really was the last operation required before the charts could be incorporated into the page. Easy really!