As January matured, my regular work load for the old reliable 9-5 began to shore up. February saw me assigned to assist on a project out of office and often out doors in freezing weather until Wednesday the 26th. A change in the schedule from five 8 hour shifts to four 9 hour shifts with a “4” on Friday (combined with a longer commute and bad weather exacerbated an already taxing configuration of circumstance), has resulted in very little being accomplished in my extracurricular projects this past month.
Returning to the office has left me with plenty of backlog and I’m attempting a series of quick sprints to catch up on all the tasks remaining for the month in a two day window. But fortune has smiled mischievously with how long the files I’m currently having to cram on to a CD-ROM by attempting to lighten the file size into PDF has given me an advantage to quickly edit some of the older posts on the Blog that didn’t get much attention as I tried to hit deadlines and live with the consequences later as opposed to not blogging at all.
The sheet hasn’t been forgotten, at every free moment I’ve had where my mind is clear and hours run side by side I’ve been working on the math. Since everything basically depends on a moon rise and moon set calculator, and how wildly inconsistent the Moon is to predict without access to extremely sophisticated algorithms that to my present knowledge are in no way public, the texts I’m working through have to do their best. And their best isn’t too far off, the one interesting thing I’ve learned since taking this project seriously was that we really didn’t have a fantastically accurate way to calculate the moon’s position ahead of time until the late 1970s and even then they were solving for noticeable discrepancies in the math up until 2005.
Luckily the way to properly manage and organize the outputs has been solved at TimeandDate.com, where with the Moon, (unlike the sun) it’s rising and setting is contained across three columns, rather than two, to accommodate the days it rises late in the day and sets sometime on the day following.
The good news is that the text I’m working for for these solves are the same text that the NOAA used to create their Excel based Sunrise calculator so some of the equasions between calculators are actually compatible now and that’s more helpful than less, however there’s a slight difference in methodologies I’ve yet to reconcile which is that the NOAA has favored converting time to Minutes, which I think is their solve for cohering the way the equations are usually calculated to how Excel usually manages time. It’s hard to explain but there’s a lot of 1440s and things being multiplied by 8 I’ve yet to understand how to marshal these fancy number techniques to my aid.
Presently there’s a computation error with the Moon’s calculated “Latitude”, and how it’s influenced by various forces in the solar system, including a set of 60 corrective calculations that adjust the result of calculating the rough or ‘mean’ latitude which is then added to the influences of Jupiter, Venus and some math that accounts to the flattening of the earth’s spherish shape, while accounting for things like the eccentricity of the earth’s orbit from moment to moment.
Between those various factors I listed above, close to 70 or 80 different equations contained in those seven categories and somewhere (probably in the argument and coefficient list) something has gone wrong by just enough to throw the equation off by .002000 of a degree. Bare in mind, 60 of these equations are super simple and repeat over and over using the same methodology. However the sheer number of them create the potential for the error to crop up and easily be missed.
This impacts the calculations that tell us the Azimuth, Altitude, Declination of the moon at a given instance. So, basically every single calculation that matters when figuring out when it will peak over the horizon of a given location, except for the bits that deal with atmospheric lensing. The calculations that give us it’s lunar ephemeris are absolutely spot on, however, and right now I can calculate very accurately the moon’s location in the zodiacal sky with extreme precision.
Topographic coordinates have been difficult to convert to, but now that I’m back at a desk during the day, I can solve for it much more quickly than in marathon bursts on the weekend.
Hopefully I’ll have some breakthrough here in the next few days and can begin figure out how to translate these new equations meaningfully into the hygromanteia sheet.
Either way, once the moonrise calculator is done and working right it’ll be released on andrewkirksey.com as it’s own downloadable file in my portfolio or on a git or something.