[GRASS-SVN] r30669 - grass/trunk/raster/r.sun
svn_grass at osgeo.org
svn_grass at osgeo.org
Fri Mar 21 06:16:52 EDT 2008
Author: neteler
Date: 2008-03-21 06:16:52 -0400 (Fri, 21 Mar 2008)
New Revision: 30669
Modified:
grass/trunk/raster/r.sun/description.html
grass/trunk/raster/r.sun/main.c
Log:
Thomas Huld <Thomas.Huld jrc.it>: bugfix: the code did not calculate the distance along the line of sight to the sun correctly, it calculated them in lat/lon distances, while the height difference was in meters! This was never correct in r.sun.
This is a workaround for lat/lon coordinates and it then REQUIRES that the elevation map is in meters. The rule is: - lat/lon coordinates: elevation in meters - Other coordinates: elevation in the same unit as the easting-northing coordinates.
Modified: grass/trunk/raster/r.sun/description.html
===================================================================
--- grass/trunk/raster/r.sun/description.html 2008-03-21 10:12:47 UTC (rev 30668)
+++ grass/trunk/raster/r.sun/description.html 2008-03-21 10:16:52 UTC (rev 30669)
@@ -10,6 +10,13 @@
for the earth curvature is internally calucated.<br>
The units of the parameters are specified in brackets, a hyphen in the brackets
explains that the parameter has no units.
+<P>
+For latitude-longitude coordinates it requires that the elevation map is in meters.
+The rules are:
+<ul>
+<li> lat/lon coordinates: elevation in meters;
+<li> Other coordinates: elevation in the same unit as the easting-northing coordinates.
+</ul>
<P>
The solar geometry of the model is based on the works of Krcho (1990), later
@@ -254,6 +261,8 @@
Marcel Suri, GeoModel, s.r.o. Bratislava, Slovakia <br>
+Thomas Huld, JRC, Italy <br>
+
© 2002, Jaroslav Hofierka, Marcel Suri
<address>
<a href="MAILTO:hofi at geomodel.sk">hofierka at geomodel.sk</a>
Modified: grass/trunk/raster/r.sun/main.c
===================================================================
--- grass/trunk/raster/r.sun/main.c 2008-03-21 10:12:47 UTC (rev 30668)
+++ grass/trunk/raster/r.sun/main.c 2008-03-21 10:16:52 UTC (rev 30669)
@@ -6,7 +6,8 @@
See manual pages for details.
(C) 2002 Copyright Jaro Hofierka, Gresaka 22, 085 01 Bardejov, Slovakia,
and GeoModel, s.r.o., Bratislava, Slovakia
-email: hofierka at geomodel.sk,marcel.suri at jrc.it,suri at geomodel.sk
+email: hofierka at geomodel.sk,marcel.suri at jrc.it,suri at geomodel.sk,
+ Thomas Huld <Thomas.Huld at jrc.it> (lat/long fix)
*******************************************************************************/
/*
* This program is free software; you can redistribute it and/or
@@ -58,6 +59,9 @@
#include <grass/gprojects.h>
#include <grass/glocale.h>
+const double deg2rad = M_PI / 180.;
+const double rad2deg = 180. / M_PI;
+
FILE *felevin, *faspin, *fslopein, *flinkein, *falbedo, *flatin;
FILE *fincidout, *fbeam_rad, *finsol_time, *fdiff_rad, *frefl_rad;
@@ -130,6 +134,23 @@
double beam_e, diff_e, refl_e, bh, dh, rr, insol_t;
double cbh, cdh;
+#define DEGREEINMETERS 111120.
+
+int ll_correction = 0;
+double coslatsq;
+
+double distance(double x1, double x2, double y1, double y2)
+{
+ if (ll_correction) {
+ return DEGREEINMETERS * sqrt(coslatsq * (x1 - x2) * (x1 - x2)
+ + (y1 - y2) * (y1 - y2));
+ }
+ else {
+ return sqrt((x1 - x2) * (x1 - x2) + (y1 - y2) * (y1 - y2));
+ }
+}
+
+
int main(int argc, char *argv[])
{
@@ -389,6 +410,10 @@
diff_rad = parm.diff_rad->answer;
refl_rad = parm.refl_rad->answer;
+ if ((G_projection() == PROJECTION_LL)) {
+ ll_correction = 1;
+ }
+
if ((insol_time != NULL) && (incidout != NULL))
G_fatal_error(_("insol_time and incidout are incompatible options"));
@@ -1159,7 +1184,7 @@
dx = (double)i *stepx;
dy = (double)j *stepy;
- length = DISTANCE1(xg0, dx, yg0, dy); /* dist from orig. grid point to the current grid point */
+ length = distance(xg0, dx, yg0, dy); /* dist from orig. grid point to the current grid point */
cube(j, i);
return;
@@ -1288,6 +1313,11 @@
func = NULL;
length = 0;
+ if (ll_correction) {
+ coslat = cos(deg2rad * yp);
+ coslatsq = coslat * coslat;
+ }
+
z_orig = z1 = zp = z[j][i];
o_orig = o[j][i];
More information about the grass-commit
mailing list