1r.resamp.rst(1) Grass User's Manual r.resamp.rst(1)
2
3
4
6 r.resamp.rst - Reinterpolates and computes topographic analysis from
7 input raster map to a new raster map (possibly with different resolu‐
8 tion) using regularized spline with tension and smoothing.
9
11 raster
12
14 r.resamp.rst
15 r.resamp.rst help
16 r.resamp.rst [-td] input=string ew_res=float ns_res=float
17 [elev=string] [slope=string] [aspect=string] [pcurv=string]
18 [tcurv=string] [mcurv=string] [smooth=string] [maskmap=string]
19 [overlap=integer] [zmult=float] [tension=float] [theta=float]
20 [scalex=float] [--overwrite] [--verbose] [--quiet]
21
22 Flags:
23 -t
24 Use dnorm independent tension
25
26 -d
27 Output partial derivatives instead of topographic parameters
28
29 --overwrite
30 Allow output files to overwrite existing files
31
32 --verbose
33 Verbose module output
34
35 --quiet
36 Quiet module output
37
38 Parameters:
39 input=string
40 Name of input raster map
41
42 ew_res=float
43 Desired east-west resolution
44
45 ns_res=float
46 Desired north-south resolution
47
48 elev=string
49 Output z-file (elevation) map
50
51 slope=string
52 Output slope map (or fx)
53
54 aspect=string
55 Output aspect map (or fy)
56
57 pcurv=string
58 Output profile curvature map (or fxx)
59
60 tcurv=string
61 Output tangential curvature map (or fyy)
62
63 mcurv=string
64 Output mean curvature map (or fxy)
65
66 smooth=string
67 Name of raster map containing smoothing
68
69 maskmap=string
70 Name of raster map to be used as mask
71
72 overlap=integer
73 Rows/columns overlap for segmentation
74 Default: 3
75
76 zmult=float
77 Multiplier for z-values
78 Default: 1.0
79
80 tension=float
81 Spline tension value
82 Default: 40.
83
84 theta=float
85 Anisotropy angle (in degrees)
86
87 scalex=float
88 Anisotropy scaling factor
89
91 r.resamp.rst reinterpolates the values a from given raster map (named
92 input) to a new raster map (named elev). Reinterpolation (resampling)
93 is done to higher, same or lower resolution specified by the ew_res and
94 ns_res parameters.
95
96 All resulting raster maps are created using the settings of the current
97 region (which may be different from that of the input raster map).
98
99 Optionally, and simultaneously with interpolation, topographic parame‐
100 ters are computed from an input raster map containing z-values of ele‐
101 vation/depth: slope, aspect, profile curvature (measured in the direc‐
102 tion of steepest slope), tangential curvature (measured in the direc‐
103 tion of a tangent to contour line) and/or mean curvature are computed
104 from and saved as raster maps as specified by the options slope,
105 aspect, pcurv, tcurv, mcurv respectively.
106
107 If the -d flag is set the program outputs partial derivatives fx, fy,
108 fxx, fxy, and fyy instead of slope, aspect and curvatures.
109
110 For noisy data it is possible to define spatially variable smoothing by
111 providing a raster map named by the smooth option containing smoothing
112 parameters. With the smoothing parameter set to zero (smooth is not
113 given or contains zero data), the resulting surface passes exactly
114 through the data points.
115
116 The user can also define a raster map (named with maskmap) which will
117 be used as a mask. The interpolation is skipped for cells which have
118 zero or NULL value in the mask.
119
120 Zero values will be assigned to these cells in all output raster maps.
121
122 The zmult parameter allows the user to rescale the z-values which may
123 be useful, e.g., for transformation of elevations given in feet to
124 meters, so that the proper values of slopes and curvatures can be com‐
125 puted. The default value is 1.
126
127 A regularized spline with tension method is used for the interpolation.
128 The tension parameter tunes the character of the resulting surface from
129 thin plate to membrane. Higher values of tension parameter reduce the
130 overshoots that can appear in surfaces with rapid change of gradient.
131
132 The -t flag can be set to use "dnorm independent tension".
133
134 The interpolation is performed for overlapping rectangular segments.
135 The user can define the width of overlap (in number of cells) with the
136 overlap option. The default value is 3.
137
139 r.resamp.rst uses regularized spline with tension for interpolation (as
140 described in Mitasova and Mitas, 1993).
141
142 The region is temporarily changed while writing output files with
143 desired resolution. Topographic parameters are computed in the same way
144 as in the v.surf.rst module. (See also Mitasova and Hofierka, 1993)
145
146 The raster map used with the smooth option should contain variable
147 smoothing parameters. These can be derived from errors, slope, etc.
148 using the r.mapcalc module.
149
150 The program gives warning when significant overshoots appear and higher
151 tension should be used. However, with tension set too high the result‐
152 ing surface changes its behavior to a membrane (rubber sheet stretched
153 over the data points resulting in a peak or pit in each given point and
154 everywhere else the surface goes rapidly to trend). Smoothing can be
155 used to reduce the overshoots. When overshoots occur the resulting elev
156 file will have white color in the locations of overshoots since the
157 color table for the output file is the same as colortable for raster
158 input file.
159
160 The program checks the numerical stability of the algorithm by computa‐
161 tion of values at given points, and prints the maximum difference found
162 into the history file of raster map elev (view with r.info). An
163 increase in tension is suggested if the difference is unacceptable.
164 For computations with smoothing set to 0 this difference should be 0.
165 With a smoothing parameter greater than zero the surface will not pass
166 through the data points exactly, and the higher the parameter the
167 closer the surface will be to the trend.
168
169 The program writes the values of parameters used in computation into
170 the comment part of the elev map history file. Additionally the follow‐
171 ing values are also written to assist in the evaluation of results and
172 choosing of suitable parameters:
173
174 minimum and maximum z values in the data file (zmin_data,
175 zmax_data) and in the interpolated raster map (zmin_int,
176 zmax_int),
177
178 maximum difference between the given and interpolated z
179 value at a given point (errtotal),
180
181 rescaling parameter used for normalization (dnorm), which
182 influences the tension.
183
184 The program gives a warning when the user wants to interpolate outside
185 the region given by the input raster map's header data. Zooming into
186 the area where the points are is suggested in this case.
187
188 When a mask is used, the program uses all points in the given region
189 for interpolation, including those in the area which is masked out, to
190 ensure proper interpolation along the border of the mask. It therefore
191 does not mask out the data points; if this is desirable, it must be
192 done outside r.resamp.rst before processing.
193
195 g.region, r.info, r.resample, r.mapcalc, r.surf.contour, v.surf.rst
196
198 Original version of program (in FORTRAN):
199 Lubos Mitas, NCSA, University of Illinois at Urbana Champaign, Il
200 Helena Mitasova, US Army CERL, Champaign, Illinois
201
202 Modified program (translated to C, adapted for GRASS , segmentation
203 procedure):
204 Irina Kosinovsky, US Army CERL.
205 Dave Gerdes, US Army CERL.
206
208 Mitas, L., Mitasova, H., 1999, Spatial Interpolation. In: P.Longley,
209 M.F. Goodchild, D.J. Maguire, D.W.Rhind (Eds.), Geographical Informa‐
210 tion Systems: Principles, Techniques, Management and Applications,
211 Wiley, 481-492.
212
213 Mitasova, H. and Mitas, L., 1993. Interpolation by regularized spline
214 with tension: I. Theory and implementation, Mathematical Geology No.25
215 p.641-656.
216
217 Mitasova, H. and Hofierka, L., 1993. Interpolation by regularized
218 spline with tension: II. Application to terrain modeling and surface
219 geometry analysis, Mathematical Geology No.25 p.657-667.
220
221 Talmi, A. and Gilat, G., 1977. Method for smooth approximation of data,
222 Journal of Computational Physics , 23, pp 93-123.
223
224 Wahba, G., 1990. Spline models for observational data, CNMS-NSF
225 Regional Conference series in applied mathematics, 59, SIAM, Philadel‐
226 phia, Pennsylvania.
227
228 Last changed: $Date: 2007-12-17 14:53:55 +0100 (Mon, 17 Dec 2007) $
229
230 Full index
231
232 © 2003-2008 GRASS Development Team
233
234
235
236GRASS 6.3.0 r.resamp.rst(1)