FW: [GMT-HELP:329] Re: how to change the color of uncoveredareausingnearneighbor
"Norman Vine" <[email protected]>
| Newsgroups | gmane.games.flightgear.terragear.devel |
|---|---|
| Message-ID | <[email protected]> |
RE: which shorelines to use ? Precision vs Resolution is an important concept that is not always fully appreciated when considering Earth data. The paper referenced below discusses one way to thin data and the results re these issues Norman > -----Original Message----- > From: [email protected] [mailto:[email protected]]On > Behalf Of Walter Smith > Sent: Wednesday, June 02, 2004 10:05 AM > To: [email protected] > Cc: [email protected] > Subject: [GMT-HELP:329] Re: how to change the color of > uncoveredareausingnearneighbor > > > Lou Estey wrote: > > > > On Mon, 31 May 2004, massimo musacchio wrote: > > > > > > The first one concerns the land mask function [grdlandmask]. There are > > > 5 kinds of resolution, how much does the highest one (FULL RESOLUTION) > > > correspond to, expressed in meters? > > > > This isn't an official answer, but based on using the it the precision of > > FULL RESOLUTION appears to be on the order of 100 meters (at least at low- > > and mid-latitudes), whereas the accuracy (how close the digital points are > > to actual ground truth) appears to vary, being as good as the precision > > in places and off by as much as several km. > > > > --lou > > Lou is correct to distinguish precision from accuracy. More clues to > answer this question may be found in Wessel and Smith's 1996 paper in JGR > on the shoreline data: > > http://www.soest.hawaii.edu/pwessel/papers/1996/JGR_96/jgr_96.html > > The first figure of that paper shows the levels of detail available in the > various resolution files used in GMT. The second figure shows that the > size-frequency distribution follows a power law for the area enclosed by a > polygon, for areas greater than about 0.1 km-squared, suggesting that the > data set captures most objects of that size and larger, but fails to > include smaller features. > > walter > > -- > Dr. Walter H. F. Smith, Geophysicist phone: 301-713-2857, ext. 126 > NOAA Lab for Satellite Altimetry, code E/RA31 fax: 301-713-3136 > 1335 East-West Highway, room 5408 mailto:[email protected] > Silver Spring MD 20910-3282 http://ibis.grdl.noaa.gov/SAT >