thesky_riset Module Reference

Procedures to determine rise, transit and set times. More...

Functions/Subroutines
subroutine	riset (jd, pl, rt, tt, st, rh, ta, sh, rsalt, for_night, ltime, cwarn, converge)
	Rise, transit and set times routine for Sun, Moon, planets and asteroids. This version recomputes positions (low accuracy first, then full accuracy) during the convergence process. See riset_ipol() for a version using interpolation, to be used for planets only(!)
subroutine	riset_ipol (jd, pl, rt, tt, st, rh, ta, sh, rsalt, for_night, ltime, cwarn)
	Old routine for rise, transit and set times for planets and asteroids - don't use this for Sun and Moon Computes three sets of planet positions, and interpolates between them.
subroutine	riset_ad (jd, ra, dec, rt, tt, st, rh, ta, sh, rsalt, for_night, cwarn)
	Rise, transit and set times routine for an object with fixed ra & dec.
subroutine	best_obs_date (jd00, ra0, m, d)
	Compute the date (m,d) at which an object with ra can be observed best, i.e., it transits at midnight
real(double) function	rsipol (y1, y2, y3, n)
	Quadratic interpolation of rise/set times for riset_ipol().

Detailed Description

Procedures to determine rise, transit and set times.

Function/Subroutine Documentation

best_obs_date()

subroutine thesky_riset::best_obs_date	(	real(double), intent(in)	jd00,
		real(double), intent(in)	ra0,
		integer, intent(out)	m,
		integer, intent(out)	d )

Compute the date (m,d) at which an object with ra can be observed best, i.e., it transits at midnight

Parameters

jd00	Julian day?
ra0	Right ascension
m	Month of best observation (output)
d	Day of month of best observation (output)

Note

• VERY SLOW!!! -> use best_obs_date_ra instead!!! (but not (yet) in libTheSky...)
• riset() should be faster now (04/2010)

Definition at line 702 of file riset.f90.

References thesky_datetime::gettz(), thesky_planetdata::nplanpos, thesky_planets::planet_position(), thesky_planetdata::planpos, riset(), and thesky_local::tz.

Here is the call graph for this function:

riset()

subroutine thesky_riset::riset	(	real(double), intent(in)	jd,
		integer, intent(in)	pl,
		real(double), intent(out)	rt,
		real(double), intent(out)	tt,
		real(double), intent(out)	st,
		real(double), intent(out)	rh,
		real(double), intent(out)	ta,
		real(double), intent(out)	sh,
		real(double), intent(in)	rsalt,
		logical, intent(in), optional	for_night,
		logical, intent(in), optional	ltime,
		logical, intent(in), optional	cwarn,
		integer, dimension(3), intent(out), optional	converge )

Rise, transit and set times routine for Sun, Moon, planets and asteroids. This version recomputes positions (low accuracy first, then full accuracy) during the convergence process. See riset_ipol() for a version using interpolation, to be used for planets only(!)

Parameters

jd	Julian day number. The rise/transit and set data are computed for the (calendar) DAY of this JD (midnight-midnight), unless for_night=.true. in which case this happens for the NIGHT starting at JD.
pl	Planet/object number (-1 - 9: ES, Moon, Mer-Plu; >10000: asteroids)
rt	Rise time (hours) (output)
tt	Transit time (hours) (output)
st	Set time (hours) (output)
rh	Rising wind direction (rad) (output)
ta	Transit altitude (rad) (output)
sh	Setting wind direction (rad) (output)
rsAlt	Altitude to return rise/set data for (degrees; 0. is actual rise/set). rsAlt>90: compute transit only
for_night	Compute rise/tranit/set data for the NIGHT starting at the indicated JD, rather than the calendar day. Times between 0h and 12h are for the NEXT calendar day! (optional; default: false)
ltime	Passed to planet_position(). If .true., include light time, doubling the CPU time while gaining a bit of accuracy (optional; default: false)
cWarn	Warn upon convergence failure (optional; default: true)
converge	Number of iterations needed to converge (optional) (output)

Note

• for rsAlt = 0.d0, rise and set times are computed
• for (rsAlt.ne.0), the routine calculates when alt=rsAlt is reached
• use rsAlt=-6,-12,-18 for the sun for twilight calculations (rsAlt is expressed in degrees).
• Returns times, rise/set azimuth and transit altitude
• Moon transit error ~0.15s (?)

Todo

• This version sometimes finds the answer tmRad(i)>2pi, which is the answer for the next day...
  • (only?) solution: use a solver on JD for az,alt

Note

Speed wrt riset_ipol(), transit only, when using low-accuracy approximations:

• Moon: 270x faster (used to be slowest by factor 3-5 wrt planets, factor 6.5 wrt Sun)
• Sun: 340x faster - full accuracy (VSOP): ~2-3x SLOWER!
• Mer: 20% slower
• Ven: 4% slower
• Mar: 10% slower
• Jup: 25% slower
• Nep: 30% slower

See also

• riset_ipol()
• Meeus, Astronomical algorithms, Ch.15, but with geographic longitude east of Greenwich defined as > 0

Definition at line 82 of file riset.f90.

References thesky_local::deltat, thesky_local::lat0, thesky_local::lon0, thesky_planets::planet_position(), thesky_planets::planet_position_la(), thesky_planetdata::planpos, thesky_coordinates::refract(), riset_ipol(), and thesky_local::tz.

Referenced by best_obs_date(), thesky_visibility::best_planet_visibility(), and thesky_visibility::planet_visibility_tonight().

Here is the call graph for this function:

riset_ad()

subroutine thesky_riset::riset_ad	(	real(double), intent(in)	jd,
		real(double), intent(in)	ra,
		real(double), intent(in)	dec,
		real(double), intent(out)	rt,
		real(double), intent(out)	tt,
		real(double), intent(out)	st,
		real(double), intent(out)	rh,
		real(double), intent(out)	ta,
		real(double), intent(out)	sh,
		real(double), intent(in)	rsalt,
		logical, intent(in), optional	for_night,
		logical, intent(in), optional	cwarn )

Rise, transit and set times routine for an object with fixed ra & dec.

Parameters

jd	Julian day number
ra	Right ascension (rad)
dec	Declination (rad)
rt	Rise time (hours) (output)
tt	Transit time (hours) (output)
st	Set time (hours) (output)
rh	Rising wind direction (rad) (output)
ta	Transit altitude (rad) (output)
sh	Setting wind direction (rad) (output)
rsAlt	Altitude to return rise/set data for (degrees; 0. is actual rise/set). rsAlt>90: compute transit only
for_night	Compute rise/tranit/set data for the NIGHT starting at the indicated JD, rather than the calendar day. Times between 0h and 12h are for the NEXT calendar day! (optional; default: false)
cWarn	Warn upon convergence failure (optional; default: true)

for rsAlt = 0., rise and set times are computed for rsAlt.ne.0, the routine calculates when alt=rsAlt is reached use rsAlt=-6,-12,-18 for the sun for twilight calculations (rsAlt is expressed in degrees).

See also

• Meeus, Astronomical algorithms, Ch.15, but with geographic longitude east of Greenwich defined as > 0

Definition at line 564 of file riset.f90.

References thesky_local::lat0, thesky_local::lon0, thesky_planets::planet_position(), thesky_planetdata::planpos, and thesky_local::tz.

Here is the call graph for this function:

riset_ipol()

subroutine thesky_riset::riset_ipol	(	real(double), intent(in)	jd,
		integer, intent(in)	pl,
		real(double), intent(out)	rt,
		real(double), intent(out)	tt,
		real(double), intent(out)	st,
		real(double), intent(out)	rh,
		real(double), intent(out)	ta,
		real(double), intent(out)	sh,
		real(double), intent(in)	rsalt,
		logical, intent(in), optional	for_night,
		logical, intent(in), optional	ltime,
		logical, intent(in), optional	cwarn )

Old routine for rise, transit and set times for planets and asteroids - don't use this for Sun and Moon Computes three sets of planet positions, and interpolates between them.

Parameters

jd	Julian day number. The rise/transit and set data are computed for the (calendar) DAY of this JD (midnight-midnight), unless for_night=.true. in which case this happens for the NIGHT starting at JD.
pl	Planet/object number (-1 - 9: ES, Moon, Mer-Plu; >10000: asteroids)
rt	Rise time (hours) (output)
tt	Transit time (hours) (output)
st	Set time (hours) (output)
rh	Rising wind direction (rad) (output)
ta	Transit altitude (rad) (output)
sh	Setting wind direction (rad) (output)
rsAlt	Altitude to return rise/set data for (degrees; 0. is actual rise/set). rsAlt>90: compute transit only
for_night	Compute rise/tranit/set data for the NIGHT starting at the indicated JD, rather than the calendar day. Times between 0h and 12h are for the NEXT calendar day! (optional; default: false)
ltime	Passed to planet_position(). If .true., include light time, doubling the CPU time while gaining a bit of accur.
cWarn	Warn upon convergence failure (optional; default: true)

Note

• for rsAlt = 0.d0, rise and set times are computed
• for (rsAlt.ne.0), the routine calculates when alt=rsAlt is reached
• use rsAlt=-6,-12,-18 for the sun for twilight calculations (rsAlt is expressed in degrees).
• transit error for Moon <= ~10s (due to interpolation); typically ~4s?
• code gets confused around midnight - it sometimes returns a small negative tmdy(i), which is on the day before(!)

See also

• Meeus, Astronomical algorithms, Ch.15, but with geographic longitude east of Greenwich defined as > 0

Definition at line 318 of file riset.f90.

References thesky_local::deltat, thesky_local::lat0, thesky_local::lon0, thesky_planets::planet_position(), thesky_planetdata::planpos, thesky_coordinates::refract(), rsipol(), and thesky_local::tz.

Referenced by riset().

Here is the call graph for this function:

rsipol()

real(double) function thesky_riset::rsipol	(	real(double), intent(in)	y1,
		real(double), intent(in)	y2,
		real(double), intent(in)	y3,
		real(double), intent(in)	n )

Quadratic interpolation of rise/set times for riset_ipol().

Parameters

y1	Coordinate for day 1
y2	Coordinate for day 2
y3	Coordinate for day 3
n	Fraction of day

Return values

rsIpol

Interpolated coordinate

Definition at line 773 of file riset.f90.

References rsipol().

Referenced by riset_ipol(), and rsipol().

Here is the call graph for this function: