HeyWhatsThat Test Sites Frequently Asked Questions
General FAQ and Site Map

This is the FAQ for a few HeyWhatsThat.com test sites.

Be sure to visit the general FAQ, which shows you all that HeyWhatsThat has to offer, and please sign up for our email announcements.

 

Test sites

Both the main site and path profiler have test sites where we're experimenting with new features.

         Main site test

         Path Profiler test

On both pages you'll find two independently adjustable profiles with lots of new options.

 

Rise/fall option on the profiler

We've received many requests, particularly from cyclists, for the total rise and fall of a path. Our answer has been "we've been told that the resolution of the SRTM elevation data doesn't give useful results."

Within the US the SRTM data gives elevation on the points of a grid whose spacing is about 100 feet, and outside the US it's 300 feet. Any bumps narrower than that spacing will not be noticed by the computation. Also, to the extent that grid points don't fall at the top of peaks and the bottom of valleys, the total rise and fall will be understated. But, to be honest, we have no idea how much that matters.

The one test we have done is to run the computation at different data resolutions. On the 11km path from 44.3°N,69.1°W to 44.4°N,69.1°W:

SRTM resolution   rise    fall
(arcseconds)
    1"            286m    220m
    3"            244m    177m
   30"             98m     54m
Higher resolution data gives more rise and fall; if you measured the path with a ruler the length of a couple of protons, you'd get really big rises and falls.

So what we've heard is that even the 1" data doesn't yield results that feel right to cyclists. But due to all your requests, we've decided to implement the feature, and now it's up to you to let us know if it's useful.

 

Setting elevation

On the main site, the viewer's position is fixed, but you can set the far end elevation in the boxes under the profiles. For the Path Profiler, you can set the elevations of any of the points along the path; click on the point's elevation (in blue on the list on the right side of the page) and enter the desired elevation.

To enter an elevation above sea level, just enter the desired elevation. The site interprets your entry as either feet or meters, according to the English/Metric setting of the page. To enter an elevation relative to ground level, use a + (plus sign) followed by the relative elevation. (For negative elevations relative to ground you'll need +-, e.g. -10 is 10 feet or meters below sea level, but +-10 is 10 feet or meters below ground level.)

 

Matching the visibility cloak

To match the results of the visibility cloak, set refraction to .14 (which implicitly sets great circle and true line of sight).

 

Profile Settings

show scale Adds horizontal and vertical scale
flat earth
curved earth
Selecting curved means we show elevations not from surface of the earth, but from a chord connecting the endpoints. This has the effect that straight lines drawn between endpoints are almost lines of sight. See the discussion in the technical FAQ.
plate caree
great circle
Previously, all profiles followed a plate carree path, a path that is linear in latitude and longitude, as described in the technical FAQ. Here you can ask for a great circle (or geodesic) path instead. (The new purple line drawn on the map is the great circle path and the brownish line is the plate carree path. You'll find you have to traverse fairly long distances to see the great circle path deviate from the linear path.)
straight line
true line of sight
Previously, the straight lines drawn on the profile were simply there to allow you to compare the heights of intervening peaks. Under the curved earth option they were fairly close to sight lines, but not exactly. Now you can ask specifically for lines of sight. Note how little difference it makes on the curved earth profiles. (true line of sight implicitly sets great circle.)
frequency Draws the first fresnel zone for the given frequency (and implicitly sets great circle and true line of sight and ignores the refraction setting). Can anyone verify this against other software?
refraction Set a value for refraction of visible light (and implicitly sets great circle and true line of sight). The default used in generating panoramas and visibility cloaks on the main site is .14, which means that the altitude of a distant peak is raised by 14% of the apparent drop it incurs due to the curvature of the earth, or, equivalently, its altitude is increased by 3.6 arc seconds for every mile distant, or, almost equivalently, that light travels in a circle whose radius is the radius of the Earth divided by .14 (e.g. 71 times larger than the earth). (I believe this is how surveyors define refraction; please let me know if there are better alternatives.)

Again, frighteningly little effect with reasonable values.

fixed exaggeration Someone requested the ability to specify a vertical/horizontal ratio. Enjoy.
y range Someone else wanted a fixed range along the Y scale. Separate the bottom and top values with a comma.
elevation data source By default we draw all our profiles with SRTM elevation data. Generated by the February 2000 Space Shuttle mission, it comprises elevations determined roughly every 100 feet north-south and east-west for the US and every 300 feet elsewhere, covering latitude 60°N to 54°S. For more information, see the NASA Jet Propulsion Laboratory and USGS Shuttle Radar Topography Mission pages.
On the profiler test page, you can ask instead for elevation from Google Maps Elevation Web Services. (Note that there's a quota on the amount data you can request through Google's service, so you may find you have to try your comparison another time. Also, we don't use the Google data to do our panorama and visibility computations, because we need hundreds of thousands of data points.)

 

Connecting the profile graph with the map

On the main site test page, you can click on the profile graph and see where on the map that particular spot is, and click on the path on the map and see where it is on the profile.

This only works if the profile is set to great circle.

When clicking on the map, if you see two path lines (which happens if your path is very long, e.g. halfway across the United States), you must click on the curved, purple line (the great circle path).

 

Airplane visibility

On the main site test page, click on the Up in the Air box in the upper right hand corner of the map to show the area where a visible airplane might be. You set the altitude in one of the colored boxes that appear below the map — you get two independent settings — and the corresponding lines show the limits where an airplane would appear above your horizon.

We do the computation by drawing sight lines past the horizon in every direction and seeing where they intersect a spherical shell at that altitude above sea level.

It's pretty rough: we don't allow altitudes below the viewer; we ignore refraction beyond the horizon; and we don't try to properly handle the case where the desired altitude is lower than the distant summit on the horizon. (eg. Suppose you're at sea level and due North there's a summit at 2,000 feet, and you ask for the visibility of planes at 1,000 feet. The right answer might be that you could see any plane between you and the summit, but our computation — which works fine if that distant summit is below 1,000 feet — will place the limit roughly halfway between you and that summit.)

By setting the profile to true line of sight and entering the altituude in far end elevation, you can click around the airplane boundary line and verify visibility (though not precisely, because as mentioned above, refraction is only partially taken into account in this computation).

 

Miscellaneous new features

  • Both pages have an input field for the number of decimal places you'd like to see in distances and bearings.

  • On the main site test page, you can choose a point on the map with the new Pan to dropdown menu in addition to clicking on the map, on the panorama, or on the list of peaks. (For example, this should make it easier to check the profiles between a set of transmitters for which you've run the panorama computations.)

  • The Profiler test page has a print option that hides most of the header material and gives you a text field where you can enter a title.

    1/2011