From: Carlie Coats on 21 Jun 2010 09:00 qquito wrote: > Dear Everyone: > > From time to time, I need to call a routine to compute the solar > zenith angle and solar azimuth angle, and I need to provide the YEAR, > MONTH and DATE and GMT as well as the LONGITUDE of the concerned > location. > > Around the end/beginning of a month and/or a year, the converting of > time involves the change of DATE, MONTH and the YEAR. It's pretty much > a nuisance! > > Is there a routine or intrinsic procedure to do the converting along > with DATE, MONTH and YEAR? > > Thank you for reading and replying! > > --Roland That's a standard kind of question in environmental modeling; one place to look is in the EPA's "SMOKE" biogenic-emissions modeling software from http://www.cmascenter.org/. Look for "biog/czangle.F" and "ioapi/daymon.f". On the other hand, you'll have to wade through a lot of stuff to get to those particular codes ;( A word of advice: the academic professors almost always phrase everything in terms of the zenith angle itself, while what you actually need is the COSINE of this angle. The straightforward spherical-trigonometry algorithm gives you the cosine directly; I've seen many codes which then take the inverse-cosine in a zenith-angle routine, and then immediately take the cosine of the result in the caller in order to use it. All they accomplish is (a) unnecessary round-off error, and (b) wasted CPU time, since trig and inverse-trig functions are computationally expensive. FWIW -- Carlie Coats
From: Philipp E. Weidmann on 21 Jun 2010 10:14 Am 20.06.2010 06:23, schrieb qquito: > Dear Everyone: > > From time to time, I need to call a routine to compute the solar > zenith angle and solar azimuth angle, and I need to provide the YEAR, > MONTH and DATE and GMT as well as the LONGITUDE of the concerned > location. > > Around the end/beginning of a month and/or a year, the converting of > time involves the change of DATE, MONTH and the YEAR. It's pretty much > a nuisance! > > Is there a routine or intrinsic procedure to do the converting along > with DATE, MONTH and YEAR? > > Thank you for reading and replying! > > --Roland > I suggest you look at http://oceancolor.gsfc.nasa.gov/cgi/idllibrary.cgi?dir=navigation which has a PROLOG source file called lonlat2solz.pro (full link http://oceancolor.gsfc.nasa.gov/DOCS/idl/navigation/lonlat2solz.pro) containing a function that does exactly what you need (after some trivial calendar conversions). It shouldn't be too hard to port this code to FORTRAN. -- -- Philipp Emanuel Weidmann
From: Arjen Markus on 22 Jun 2010 02:56
On 20 jun, 19:56, Arjan <arjan.van.d...(a)rivm.nl> wrote: > > From time to time, I need to call a routine to compute the solar > > zenith angle and solar azimuth angle, and I need to provide the YEAR, > > MONTH and DATE and GMT as well as the LONGITUDE of the concerned > > location. > > You can use my routines for solar angles (plus the ones for date/ > time). > Send me an e-mail at > arjan [dot] van [dot] dijk [at] rivm [dot] nl. > > Regards, > > Arjan The routines I referred are by Arjan by the way ;). Regards, Arj_e_n |