Bearing to Yerushalayim and Zmanim Map 2.0

Vintage Map with CompassThe availability of the Bearing to Yerushalayim and Zmanim Map was originally announced on December 30, 2007. At the time there were a number of bugs related to the Google Map API. These bugs were reported to Google and eventually fixed. Since that time, the only change was a minor JavaScript fix for IE. The Bearing to Yerushalayim worked, but the zmanim tabs had a major issue because the timezone calculated was done based on the user’s current browser timezone. This made it tricky to check zmanim in a different location or timezone than the user’s current timezone.Zmanim tab using timezonesI recently updated the map to look-up the actual timezone of the latitude and longitude selected by the user. This was implemented by doing a look-up at the geonames.org timezone web-service. The timezone is passed to the Zmanim API and used to generate the XML output of a list of daily zmanim that is displayed in the map. Since the Olson timezone database changes a few times a year, there will almost certainly be cases where the proper timezone can’t be determined. Some of these are changes of timezone names, such as the change from Asia/Calcutta to Asia/Kolkata (my host will not run the TZ Updater tool). In these cases, a simple mapping between the old and new was added to the map. In cases where the timezone can’t be determined the timezone will default to GMT. Ocean locations within 10 km of land will use the closest landmass, but anywhere beyond 10 km will default to GMT. One issue with using the geonames.org webservice, is that when it is down, the map will timeout. I experimented with various ways of dealing with this, but unless my host updates the Java version from 1.4, they are too complex to use at this time.

See the Technical Information about the Bearing to Yerushalayim Map post for technical details about the original implementation.

FAQ: How Much Earlier is Sunrise on Mount Everest Due to Elevation?

Sunrise over Everest

Question:

How Much Earlier is Sunrise on Mount Everest Due to Elevation?

Answer:

The greatest sunrise and sunset elevation effect on Earth is on Mount Everest (at 27.988° N, 86.925° E as seen on the Direction to Yerushalayim Map). With an elevation of 8,848 Meters (29, 029 feet), sunrise would be up to 15 minutes and 31 seconds earlier on Mount Everest than at sea level. The range of the effect is from 15 minutes and 31 seconds on June 22nd, to a “low” of 13 minutes 41 seconds earlier on March 18th. Being in a large mountain range with obstructed horizons, it is likely never actually seen that early. In addition to questions about mountains, every few months I get asked about how much earlier sunrise/set can be seen in skyscrapers. There are various halacha questions as to whether this actually affects zmanim that I will mention later. Here are some raw numbers. Burj KhalifaBurj Khalifa (at 25.197° N, 55.274° E as seen on the Direction to Yerushalayim Map) is the tallest building in the world. With a height of 828 m (2,717 ft), visible sunrise to someone standing on top of the crown (something unrealistic) on June 22nd would be at 5:24:56 AM versus 5:29:31 AM on sea level, a difference of 4 minutes and 35 seconds. Sunset would be 7:16:35 PM versus 7:12:00 PM at sea level, a difference of 4 minutes and 35 seconds. A more realistic scenario would be the visibility sunrise on the highest floor (the 160th), an elevation of 672 m (2,204 ft) at 5:25:23 AM, a difference of 4 minutes and 8 seconds earlier than sea level. Sunset on the 160th floor would be 7:16:08 PM, or 4 minutes and 8 seconds later than at sea level.
As far as the halacha being affected by the elevation of buildings, the Baal Hatanya seems to indicate that tall buildings would make sunset later. See Yisroel vehazmanim ישראל והזמנים Vol II, page 910. In the Shraga Lachaim שרגא לחיים footnotes Rabbi Harfenes states that

ויש להוסיף שהוא דבר תמוה לומר דעד שלא נבנו הבנינים הגבוהים היה זמן שבת התלוי בשקעה”ח (שקיעה ראשונה להגאנים ושקיעה שניה לר”ת) מוקדם, ולאחר שנבנו יש לאחר הזמנים, ועד עכשיו שהיו בניו יארק הבנינים התאומויות (טווין-טאוע”ר בלע”ז) שכל א’ מהם היה בת ק”י קומות היה זמן השקיעה מאוחר, ועתה לאחר שהפילו והרסו אותם רשעים וזדים ארורים ימ”ש חזר הדבר לקדמותו להקדים זמן השקיעה.

Zmanim API 1.2 Released

The Zmanim API 1.2 was released today. The only change in this release is the removal of elevation as a factor in the calculation of all zmanim besides sunrise and sunset. Updated JavaDocs now reflect these changes. Please see the Elevation Now Only Impacts Sunrise and Sunset Calculations in the Zmanim API post for additional information.

I would like to thank the various people who contacted me on the subject.

The main download is the Zmanim 1.2 release zip file that includes source files and JavaDoc documentation. Also available for download (included in the above zip file) is the main zmanim-1.2.jar. The zmanimAstronomical-1.1.jar that only includes the AstronomicalCalendar was not impacted by this change, but the version has been updated to zmanimAstronomical-1.2.jar for consistency. Additional detail on the downloads can be seen on the Zmanim Download page

Elevation Now Only Impacts Sunrise and Sunset Calculations in the Zmanim API

Sunrise over Everest

Note: On Aug 1, 2018 – כ״א אב תשע״ח, elevation became optionally available as a factor in most zmanim calculations. The KosherJava Zmanim library was updated to optionally allow using elevation as an attribute that will adjust most zmanim via the ZmanimCalendar​.setUseElevation(boolean). There are some zmanim such as degree-based zmanim that are driven by the amount of light in the sky and are not impacted by elevation. These zmanim intentionally do not support elevation adjustment. The following is the original post.


I have received a few emails over the past year or so about the use of elevation as part of the calculation of various zmanim in the API. All releases of the zmanim API up to the current 1.1 release, use elevation to affect the calculation of all sunrise/sunset based zmanim. While my footnote on the Zmanim Calendar Generator page stated that this only impacted sunrise and sunset (quoting the Yisrael Vehazmanim and Calendrical Calculations), in practice all zmanim calculated based on sunrise and sunset were affected by elevation. An example of a zman that used elevation in 1.1, but will no longer in 1.2, is the calculation of Sof Zman Krias Shema according to the GR”A. The soon to be released version 1.2 limits the impact of elevation to sunrise and sunset only. The Zmanim Calendar Generator page already uses the latest changes to the code. To facilitate changes by ports of the API, a list of changes to the source code of the ZmanimCalendar.java and ComplexZmanimCalendar.java is available in a diff file (a proper Unix diff/patch is not available at this time). The diff files were generated prior to updating the JavaDocs to simplify the work of anyone modifying ports of the Zmanim API. Detailed sources for not using elevation for anything besides sunrise and sunset can be found in פרק ב׳ סעיף ט׳ (pages 186-187) in the second edition of Zmanim Kehilchasam זמנים כהלכתם by Rabbi Dovid Yehuda Bursztyn.

Calculating the Bearing/Direction to Har Habayis Using the Zmanim API

Vintage Map with CompassAn earlier “Bearing to Yerushalayim and Zmanim Map” post demonstrated the use of JavaScript to render the bearing to Har Habayis on a Google Map. A more detailed follow-up post “Technical Information about the Bearing to Yerushalayim Map” dealt with detailed technical information on these calculations. The main Bearing to Yerushalayim and Zmanim Map page usually has the most up to date information on the subject. What was not detailed in previously published posts and pages was that most of the calculations available via JavaScript are now in the core Zmanim API. Available since the July, 2008 beta 2 release of version 1.1 is the ability to bearings/directions using both the great circle and rhumb line methods in Java. The GeoLocation Object was modified to calculate the great circle bearings (both initial and final), and rhumb line bearing from any GeoLocation Object to another. In addition, distance calculation between the two points using both of these line types is supported. What was not ported from the JavaScript version was the less accurate Haversine formula, or the simpler spherical law of cosines algorithms that yield identical results. Instead, the Zmanim API uses the far more accurate Vincenty formulae using the WGS84 geoid model of the earth. Published by the geodesist/mathematician Thaddeus Vincenty, it is said to be accurate to about one-half millimeter, more than adequate for our calculation. The code in the API is a Java port of the previously published, slightly modified version of Chris Veness’s JavaScript implementation . Below is a simple Java example of generating bearing and distances.

/**
 * This program demonstrates how to calculate bearing to Yerushalayim
 * using the kosherjava.com Zmanim API. Both the great circle and
 * rhumb line method are shown
 * To compile, ensure that the Zmanim Jar is in your classpath.
 */
import com.kosherjava.zmanim.util.GeoLocation;
import java.util.TimeZone;

public class BearingToYerushalayim{
	public static void main(String [] args) {
		GeoLocation lakewood = new GeoLocation("Lakewood, NJ", 40.09596, -74.22213, 0, TimeZone.getTimeZone("America/New_York"));
		GeoLocation harHabayis = new GeoLocation("Har Habayis", 31.77805, 35.235149, 0, TimeZone.getTimeZone("Asia/Jerusalem"));

		double greatCircleInitialBearing = lakewood.getGeodesicInitialBearing(harHabayis);
		double greatCircleDistance = lakewood.getGeodesicDistance(harHabayis);

		double rhumbLineBearing = lakewood.getRhumbLineBearing(harHabayis);
		double rhumbLineDistance = lakewood.getRhumbLineDistance(harHabayis);

		System.out.println("Great circle initial bearing: " + greatCircleInitialBearing + " degrees ");
		System.out.println("Great circle distance: " + greatCircleDistance / 1000 + " km");

		System.out.println("Rhumb line bearing: " + rhumbLineBearing + " degrees");
		System.out.println("Rhumb line distance: " + rhumbLineDistance / 1000 + " km");

	}
}