1 | ! $Id: ropp_pp_preprocess_GRASRS.f90 2021 2009-01-16 10:49:04Z frhl $
|
---|
2 |
|
---|
3 | SUBROUTINE ropp_pp_preprocess_GRASRS(ro_data, config, LCF)
|
---|
4 |
|
---|
5 | !****s* Preprocessing/ropp_pp_preprocess_GRASRS *
|
---|
6 | !
|
---|
7 | ! NAME
|
---|
8 | ! ropp_pp_preprocess_GRASRS - Mission-specific Level1a data preprocessing
|
---|
9 | ! for GRAS Raw Sampling data
|
---|
10 | !
|
---|
11 | ! SYNOPSIS
|
---|
12 | ! call ropp_pp_preprocess_GRASRS(ro_data, config, LCF)
|
---|
13 | !
|
---|
14 | ! DESCRIPTION
|
---|
15 | ! Merge and upsample CL and RS data
|
---|
16 | ! 1. Select CL and RS records by LCF flag
|
---|
17 | ! 2. Generate merged time grid anchored at highest point of RS record
|
---|
18 | ! 3. Interpolate CL and RS data to merged time grid
|
---|
19 | ! 4. Merge data
|
---|
20 | ! 5. Generate phase model and connecting phase
|
---|
21 | ! 6. Interpolate data on merged time grid within small gaps
|
---|
22 | ! 7. Restore phase variation
|
---|
23 | !
|
---|
24 | ! INPUTS
|
---|
25 | ! type(ROprof) :: ro_data ! Radio occultation data strucuture
|
---|
26 | ! type(PPConfig) :: config ! Configuration options
|
---|
27 | ! integer :: LCF ! Lost carrier flag
|
---|
28 | !
|
---|
29 | ! OUTPUT
|
---|
30 | ! type(ROprof) :: ro_data ! Corrected radio occultation data
|
---|
31 | ! type(PPConfig) :: config ! Configuration options
|
---|
32 | ! integer :: LCF ! Lost carrier flag
|
---|
33 | !
|
---|
34 | ! NOTES
|
---|
35 | ! Requires ROprof data structure type, defined in ropp_io module. This
|
---|
36 | ! routine therefore requires that the ropp_io module is pre-installed before
|
---|
37 | ! compilation.
|
---|
38 | !
|
---|
39 | ! REFERENCES
|
---|
40 | !
|
---|
41 | ! AUTHOR
|
---|
42 | ! M Gorbunov, Russian Academy of Sciences, Russia.
|
---|
43 | ! Any comments on this software should be given via the ROM SAF
|
---|
44 | ! Helpdesk at http://www.romsaf.org
|
---|
45 | !
|
---|
46 | ! COPYRIGHT
|
---|
47 | ! Copyright (c) 1998-2010 Michael Gorbunov <michael.gorbunov@zmaw.de>
|
---|
48 | ! For further details please refer to the file COPYRIGHT
|
---|
49 | ! which you should have received as part of this distribution.
|
---|
50 | !
|
---|
51 | !****
|
---|
52 |
|
---|
53 | !-------------------------------------------------------------------------------
|
---|
54 | ! 1. Declarations
|
---|
55 | !-------------------------------------------------------------------------------
|
---|
56 |
|
---|
57 | USE typesizes, ONLY: wp => EightByteReal
|
---|
58 | USE arrays, ONLY: WHERE
|
---|
59 | USE ropp_utils, ONLY: impact_parameter, ropp_MDFV, occ_point
|
---|
60 | USE ropp_io, ONLY: ropp_io_init, ropp_io_free, ropp_io_addvar
|
---|
61 | USE ropp_io_types, ONLY: ROprof, L1atype
|
---|
62 | USE ropp_pp_preproc, not_this => ropp_pp_preprocess_GRASRS
|
---|
63 | USE ropp_pp
|
---|
64 | USE ropp_pp_spline
|
---|
65 | USE ropp_pp_constants
|
---|
66 | USE ropp_pp_types, ONLY: PPConfig
|
---|
67 |
|
---|
68 | IMPLICIT NONE
|
---|
69 |
|
---|
70 | TYPE(ROprof), INTENT(inout) :: ro_data ! Occultation data struct
|
---|
71 | TYPE(PPconfig), INTENT(inout) :: config ! Configuration options
|
---|
72 | INTEGER, DIMENSION(:), POINTER :: LCF ! Lost carrier flag
|
---|
73 |
|
---|
74 | INTEGER, PARAMETER :: np = 500 ! Dimension phase model
|
---|
75 | INTEGER, PARAMETER :: nt = 300 ! RS grid interval
|
---|
76 | INTEGER, PARAMETER :: nv = 5 ! Regression order
|
---|
77 | ! REAL(wp), PARAMETER :: gap = 0.8 ! Maximum data gap size
|
---|
78 | REAL(wp), PARAMETER :: gap = 0.04 ! Maximum data gap size
|
---|
79 |
|
---|
80 | TYPE(L1atype) :: mg_data ! Merged data
|
---|
81 | TYPE(L1atype) :: rs_data ! Raw sampling data
|
---|
82 | TYPE(L1atype) :: cl_data ! Closed loop data
|
---|
83 | INTEGER, DIMENSION(:), ALLOCATABLE :: mg_lcf ! Merged LCF
|
---|
84 | INTEGER, DIMENSION(:), ALLOCATABLE :: rs_lcf ! Raw sampling LCF
|
---|
85 | INTEGER, DIMENSION(:), ALLOCATABLE :: cl_lcf ! Closed loop LCF
|
---|
86 |
|
---|
87 | REAL(wp), DIMENSION(:), ALLOCATABLE :: mg_msis ! MSIS phase model (merged)
|
---|
88 | REAL(wp), DIMENSION(:), ALLOCATABLE :: mg_impact ! MSIS phase model (merged)
|
---|
89 | REAL(wp), DIMENSION(:), ALLOCATABLE :: mg_ds ! phase deviation (merged)
|
---|
90 | REAL(wp), DIMENSION(:), ALLOCATABLE :: t_norm ! Normalised time
|
---|
91 | REAL(wp), DIMENSION(:,:),ALLOCATABLE :: KV ! Regression matrix
|
---|
92 | REAL(wp), DIMENSION(0:nv,3) :: coeff_vleo ! Regression coeffs
|
---|
93 | REAL(wp), DIMENSION(0:nv,3) :: coeff_vgns ! Regression coeffs
|
---|
94 |
|
---|
95 | INTEGER :: ocd
|
---|
96 | INTEGER :: i, j, n, n1, n_cl, n_rs
|
---|
97 | INTEGER :: icl_min, icl_max
|
---|
98 | INTEGER :: irs_min, irs_max
|
---|
99 | INTEGER :: ig1, ig2, igl
|
---|
100 | INTEGER :: i_int
|
---|
101 | LOGICAL :: mrg_Mode ! Merging mode
|
---|
102 |
|
---|
103 | REAL(wp) :: p1, pN, sb
|
---|
104 | REAL(wp) :: ts, t1, tgl, tmin, tmax
|
---|
105 |
|
---|
106 | INTEGER, DIMENSION(:), POINTER :: idx
|
---|
107 | INTEGER :: nidx
|
---|
108 | REAL(wp), DIMENSION(:,:), ALLOCATABLE :: dtime_idc
|
---|
109 |
|
---|
110 | !-------------------------------------------------------------------------------
|
---|
111 | ! 2. Retrieve lost carrier flag information from input data file
|
---|
112 | !-------------------------------------------------------------------------------
|
---|
113 |
|
---|
114 | IF (ASSOCIATED(ro_data%vlist%VlistD1d)) THEN
|
---|
115 | LCF(:) = NINT(ro_data%vlist%VlistD1d%data(1:size(LCF)))
|
---|
116 | DEALLOCATE(ro_data%vlist%VlistD1d%data)
|
---|
117 | ENDIF
|
---|
118 |
|
---|
119 | !-------------------------------------------------------------------------------
|
---|
120 | ! 3. Data quality checks
|
---|
121 | !-------------------------------------------------------------------------------
|
---|
122 |
|
---|
123 | WHERE(ro_data%Lev1a%phase_L1(:) == ropp_MDFV)
|
---|
124 | ro_data%Lev1a%phase_L1(:) = -1.0_wp
|
---|
125 | ENDWHERE
|
---|
126 | WHERE(ro_data%Lev1a%phase_L2(:) == ropp_MDFV)
|
---|
127 | ro_data%Lev1a%phase_L2(:) = -1.0_wp
|
---|
128 | ENDWHERE
|
---|
129 |
|
---|
130 | !-------------------------------------------------------------------------------
|
---|
131 | ! 4. Initialisation
|
---|
132 | !-------------------------------------------------------------------------------
|
---|
133 |
|
---|
134 | mrg_Mode = .TRUE.
|
---|
135 |
|
---|
136 | ! 4.1 Select CL data
|
---|
137 |
|
---|
138 | icl_min = SUM(MINLOC(ro_data%lev1a%dtime(:), MASK = .NOT. BTEST(LCF(:),0)))
|
---|
139 | icl_max = SUM(MAXLOC(ro_data%lev1a%dtime(:), MASK = .NOT. BTEST(LCF(:),0)))
|
---|
140 |
|
---|
141 | n_cl = icl_max-icl_min+1
|
---|
142 | CALL ropp_io_init(cl_data, n_cl)
|
---|
143 | ALLOCATE(cl_lcf(n_cl))
|
---|
144 |
|
---|
145 | cl_data%dtime = ro_data%lev1a%dtime(icl_min:icl_max)
|
---|
146 | DO j=1,3
|
---|
147 | cl_data%r_gns(:,j) = ro_data%lev1a%r_gns(icl_min:icl_max,j)
|
---|
148 | cl_data%v_gns(:,j) = ro_data%lev1a%v_gns(icl_min:icl_max,j)
|
---|
149 | cl_data%r_leo(:,j) = ro_data%lev1a%r_leo(icl_min:icl_max,j)
|
---|
150 | cl_data%v_leo(:,j) = ro_data%lev1a%v_leo(icl_min:icl_max,j)
|
---|
151 | ENDDO
|
---|
152 | cl_data%snr_L1ca = ro_data%lev1a%snr_L1ca(icl_min:icl_max)
|
---|
153 | cl_data%snr_L1p = ro_data%lev1a%snr_L1p(icl_min:icl_max)
|
---|
154 | cl_data%snr_L2p = ro_data%lev1a%snr_L2p(icl_min:icl_max)
|
---|
155 | cl_data%phase_L1 = ro_data%lev1a%phase_L1(icl_min:icl_max)
|
---|
156 | cl_data%phase_L2 = ro_data%lev1a%phase_L2(icl_min:icl_max)
|
---|
157 | cl_lcf = lcf(icl_min:icl_max)
|
---|
158 |
|
---|
159 | ! 4.2 Select RS data
|
---|
160 |
|
---|
161 | irs_min = SUM(MINLOC(ro_data%lev1a%dtime(:), MASK = BTEST(LCF(:),0)))
|
---|
162 | ! irs_min = SUM(MINLOC(ro_data%lev1a%dtime(:), MASK = (BTEST(LCF(:),0).and..not.BTEST(LCF(:),3))))
|
---|
163 | irs_max = SUM(MAXLOC(ro_data%lev1a%dtime(:), MASK = BTEST(LCF(:),0)))
|
---|
164 | ! irs_max = SUM(MAXLOC(ro_data%lev1a%dtime(:), MASK = (BTEST(LCF(:),0).and..not.BTEST(LCF(:),3))))
|
---|
165 |
|
---|
166 | IF (irs_min >= 1 .AND. irs_min <= ro_data%lev1a%npoints) THEN
|
---|
167 |
|
---|
168 | n_rs = irs_max-irs_min+1
|
---|
169 | CALL ropp_io_init(rs_data, n_rs)
|
---|
170 | ALLOCATE(rs_lcf(n_rs))
|
---|
171 | rs_data%dtime = ro_data%lev1a%dtime(irs_min:irs_max)
|
---|
172 | DO j=1,3
|
---|
173 | rs_data%r_gns(:,j) = ro_data%lev1a%r_gns(irs_min:irs_max,j)
|
---|
174 | rs_data%v_gns(:,j) = ro_data%lev1a%v_gns(irs_min:irs_max,j)
|
---|
175 | rs_data%r_leo(:,j) = ro_data%lev1a%r_leo(irs_min:irs_max,j)
|
---|
176 | rs_data%v_leo(:,j) = ro_data%lev1a%v_leo(irs_min:irs_max,j)
|
---|
177 | ENDDO
|
---|
178 | rs_data%snr_L1ca = ro_data%lev1a%snr_L1ca(irs_min:irs_max)
|
---|
179 | rs_data%snr_L1p = ro_data%lev1a%snr_L1p(irs_min:irs_max)
|
---|
180 | rs_data%snr_L2p = ro_data%lev1a%snr_L2p(irs_min:irs_max)
|
---|
181 | rs_data%phase_L1 = ro_data%lev1a%phase_L1(irs_min:irs_max)
|
---|
182 | rs_data%phase_L2 = ro_data%lev1a%phase_L2(irs_min:irs_max)
|
---|
183 | rs_lcf = lcf(irs_min:irs_max)
|
---|
184 |
|
---|
185 |
|
---|
186 | ! idx => WHERE((BTEST(LCF(:),0)) .AND. (.NOT.BTEST(LCF(:),3)), nidx)
|
---|
187 |
|
---|
188 |
|
---|
189 | ! IF (nidx > 0) THEN
|
---|
190 |
|
---|
191 | ! n_rs = nidx
|
---|
192 | ! CALL ropp_io_init(rs_data, n_rs)
|
---|
193 | ! ALLOCATE(rs_lcf(n_rs))
|
---|
194 | ! rs_data%dtime = ro_data%lev1a%dtime(idx)
|
---|
195 | ! DO j=1,3
|
---|
196 | ! rs_data%r_gns(:,j) = ro_data%lev1a%r_gns(idx,j)
|
---|
197 | ! rs_data%v_gns(:,j) = ro_data%lev1a%v_gns(idx,j)
|
---|
198 | ! rs_data%r_leo(:,j) = ro_data%lev1a%r_leo(idx,j)
|
---|
199 | ! rs_data%v_leo(:,j) = ro_data%lev1a%v_leo(idx,j)
|
---|
200 | ! END DO
|
---|
201 | ! rs_data%snr_L1ca = ro_data%lev1a%snr_L1ca(idx)
|
---|
202 | ! rs_data%snr_L1p = ro_data%lev1a%snr_L1p(idx)
|
---|
203 | ! rs_data%snr_L2p = ro_data%lev1a%snr_L2p(idx)
|
---|
204 | ! rs_data%phase_L1 = ro_data%lev1a%phase_L1(idx)
|
---|
205 | ! rs_data%phase_L2 = ro_data%lev1a%phase_L2(idx)
|
---|
206 | ! rs_lcf = lcf(idx)
|
---|
207 |
|
---|
208 | ELSE
|
---|
209 |
|
---|
210 | n_rs = 0
|
---|
211 | mrg_Mode = .FALSE. ! Closed loop data only
|
---|
212 |
|
---|
213 | ENDIF
|
---|
214 |
|
---|
215 | !-------------------------------------------------------------------------------
|
---|
216 | ! 5. Generate merged time grid
|
---|
217 | !-------------------------------------------------------------------------------
|
---|
218 |
|
---|
219 | ! 5.1 Determine occultation direction
|
---|
220 |
|
---|
221 | p1 = impact_parameter(cl_data%r_leo(1,:), cl_data%r_gns(1,:))
|
---|
222 | pN = impact_parameter(cl_data%r_leo(n_cl,:), cl_data%r_gns(n_cl,:))
|
---|
223 | ocd = NINT(SIGN(1.0_wp, pN - p1))
|
---|
224 |
|
---|
225 | ! 5.2 Determine reference points and grid time step
|
---|
226 |
|
---|
227 | IF (mrg_Mode) THEN ! Merge RS+CL
|
---|
228 |
|
---|
229 | ts = MINVAL(ABS(rs_data%dtime(1+nt:n_rs)-rs_data%dtime(1:n_rs-nt)))/nt
|
---|
230 | IF (ocd < 0) THEN
|
---|
231 | tmin = MINVAL(cl_data%dtime(:))
|
---|
232 | tmax = MAXVAL(rs_data%dtime(:))
|
---|
233 | t1 = MINVAL(rs_data%dtime(:))
|
---|
234 | ELSE
|
---|
235 | tmin = MINVAL(rs_data%dtime(:))
|
---|
236 | tmax = MAXVAL(cl_data%dtime(:))
|
---|
237 | t1 = MAXVAL(rs_data%dtime(:))
|
---|
238 | ENDIF
|
---|
239 |
|
---|
240 | ELSE
|
---|
241 |
|
---|
242 | ts = MINVAL(ABS(cl_data%dtime(2:n_cl)-cl_data%dtime(1:n_cl-1)))
|
---|
243 | tmin = MINVAL(cl_data%dtime(:))
|
---|
244 | tmax = MAXVAL(cl_data%dtime(:))
|
---|
245 | IF (ocd < 0) THEN
|
---|
246 | t1 = MINVAL(cl_data%dtime(:))
|
---|
247 | ELSE
|
---|
248 | t1 = MAXVAL(cl_data%dtime(:))
|
---|
249 | ENDIF
|
---|
250 |
|
---|
251 | ENDIF
|
---|
252 |
|
---|
253 | ! 5.3 Determine anchor point index and grid dimension
|
---|
254 |
|
---|
255 | n1 = FLOOR((t1 - tmin)/ts) + 1
|
---|
256 | n = FLOOR((tmax - t1)/ts) + n1
|
---|
257 |
|
---|
258 | ! 5.4 Generate merged time grid
|
---|
259 | CALL ropp_io_init(mg_data, n)
|
---|
260 | ALLOCATE(mg_LCF(n))
|
---|
261 | mg_data%reference_frame%r_leo = ro_data%lev1a%reference_frame%r_leo
|
---|
262 | mg_data%reference_frame%r_gns = ro_data%lev1a%reference_frame%r_gns
|
---|
263 |
|
---|
264 | IF (ocd < 0) THEN
|
---|
265 | DO i=1,n
|
---|
266 | mg_data%dtime(i) = ((n-i)*t1 + (i-n1)*tmax)/(n-n1)
|
---|
267 | ENDDO
|
---|
268 | ELSE
|
---|
269 | DO i=1,n
|
---|
270 | mg_data%dtime(i) = ((n1-i)*tmin + (i-1)*t1)/(n1-1)
|
---|
271 | ENDDO
|
---|
272 | ENDIF
|
---|
273 | print*,'n, minval(mg_data%dtime), maxval(mg_data%dtime), ro_data%Lev1a%npoints = ', n, minval(mg_data%dtime), maxval(mg_data%dtime), ro_data%Lev1a%npoints
|
---|
274 |
|
---|
275 | !-------------------------------------------------------------------------------
|
---|
276 | ! 6. Interpolate CL and RS data to merged time grid
|
---|
277 | !-------------------------------------------------------------------------------
|
---|
278 |
|
---|
279 | ! 6.1 Interpolate trajectories
|
---|
280 |
|
---|
281 | ALLOCATE(t_norm(ro_data%Lev1a%npoints))
|
---|
282 | ! ALLOCATE(t_norm(n_cl+n_rs))
|
---|
283 | ALLOCATE(KV(ro_data%Lev1a%npoints, 0:nv))
|
---|
284 | ! ALLOCATE(KV(n_cl+n_rs, 0:nv))
|
---|
285 | t_norm(:) = (ro_data%Lev1a%dtime(:) - ro_data%Lev1a%dtime(1))/ &
|
---|
286 | (ro_data%Lev1a%dtime(ro_data%Lev1a%npoints) - ro_data%Lev1a%dtime(1))
|
---|
287 | ! t_norm(:) = (ro_data%Lev1a%dtime(:n_cl+n_rs) - ro_data%Lev1a%dtime(1))/ &
|
---|
288 | ! (ro_data%Lev1a%dtime(n_cl+n_rs) - ro_data%Lev1a%dtime(1))
|
---|
289 | CALL ropp_pp_init_polynomial(t_norm, KV)
|
---|
290 | DO j=1,3
|
---|
291 | CALL ropp_pp_regression(KV,ro_data%Lev1a%r_leo(:,j),coeff_vleo(:,j))
|
---|
292 | ! CALL ropp_pp_regression(KV,ro_data%Lev1a%r_leo(:n_cl+n_rs,j),coeff_vleo(:,j))
|
---|
293 | CALL ropp_pp_regression(KV,ro_data%Lev1a%r_gns(:,j),coeff_vgns(:,j))
|
---|
294 | ! CALL ropp_pp_regression(KV,ro_data%Lev1a%r_gns(:n_cl+n_rs,j),coeff_vgns(:,j))
|
---|
295 | ENDDO
|
---|
296 |
|
---|
297 | ! For diagnostics, plot out residuals evaluated at original times.
|
---|
298 |
|
---|
299 | DO i=1,ro_data%Lev1a%npoints
|
---|
300 | CALL ropp_pp_interpolate_trajectory(ro_data%Lev1a%dtime, &
|
---|
301 | coeff_vleo, coeff_vgns, &
|
---|
302 | ro_data%georef%r_coc, &
|
---|
303 | ro_data%Lev1a%dtime(i), &
|
---|
304 | mg_data%r_leo(i,:), &
|
---|
305 | mg_data%v_leo(i,:), &
|
---|
306 | mg_data%r_gns(i,:), &
|
---|
307 | mg_data%v_gns(i,:))
|
---|
308 | ENDDO
|
---|
309 |
|
---|
310 | print*, 'SIZE(LCF) = ', SIZE(LCF)
|
---|
311 | print*, 'LCF(1:5) = ', LCF(1:5)
|
---|
312 |
|
---|
313 | ALLOCATE(dtime_idc(ro_data%Lev1a%npoints, 3))
|
---|
314 | dtime_idc(:, 1) = mg_data%dtime(1:ro_data%Lev1a%npoints)
|
---|
315 | dtime_idc(:, 2) = ro_data%Lev1a%dtime(1:ro_data%Lev1a%npoints)
|
---|
316 | dtime_idc(:, 3) = REAL(LCF, wp)
|
---|
317 | CALL ropp_io_addvar( &
|
---|
318 | ro_data, &
|
---|
319 | name = "dtime_idc", &
|
---|
320 | long_name = "mg_data dtime, ro_data dtime and lcf", &
|
---|
321 | units = "s", &
|
---|
322 | range = (/-1.0E6_wp, 1.0E6_wp/), &
|
---|
323 | DATA = dtime_idc)
|
---|
324 | DEALLOCATE(dtime_idc)
|
---|
325 |
|
---|
326 | print*,'SHAPE(ro_data%Lev1a%r_gns) = ', SHAPE(ro_data%Lev1a%r_gns)
|
---|
327 | print*,'SHAPE(mg_data%r_gns) = ', SHAPE(mg_data%r_gns)
|
---|
328 | print*,'SHAPE(mg_data%r_gns-ro_data%Lev1a%r_gns) = ', SHAPE(mg_data%r_gns-ro_data%Lev1a%r_gns)
|
---|
329 |
|
---|
330 | print*,'SHAPE(ropp_io_isnan(ro_data%Lev1a%r_gns)) = ', SHAPE(ropp_io_isnan(ro_data%Lev1a%r_gns))
|
---|
331 | print*,'COUNT(ropp_io_isnan(ro_data%Lev1a%r_gns)) = ', COUNT(ropp_io_isnan(ro_data%Lev1a%r_gns))
|
---|
332 |
|
---|
333 | print*,'SHAPE(ropp_io_isnan(mg_data%r_gns)) = ', SHAPE(ropp_io_isnan(mg_data%r_gns))
|
---|
334 | print*,'COUNT(ropp_io_isnan(mg_data%r_gns)) = ', COUNT(ropp_io_isnan(mg_data%r_gns))
|
---|
335 |
|
---|
336 | print*,'SHAPE(ropp_io_isnan(ro_data%Lev1a%r_gns) .OR. ropp_io_isnan(mg_data%r_gns)) = ', SHAPE(ropp_io_isnan(ro_data%Lev1a%r_gns) .OR. ropp_io_isnan(mg_data%r_gns))
|
---|
337 | print*,'COUNT(ropp_io_isnan(ro_data%Lev1a%r_gns) .OR. ropp_io_isnan(mg_data%r_gns)) = ', COUNT(ropp_io_isnan(ro_data%Lev1a%r_gns) .OR. ropp_io_isnan(mg_data%r_gns))
|
---|
338 |
|
---|
339 | ! WHERE (ropp_io_isnan(ro_data%Lev1a%r_gns(1:ro_data%Lev1a%npoints)) .OR. ropp_io_isnan(mg_data%r_gns(1:ro_data%Lev1a%npoints)))
|
---|
340 | ! mg_data%r_gns = ropp_MDFV
|
---|
341 | ! ELSEWHERE
|
---|
342 | ! mg_data%r_gns = ro_data%Lev1a%r_gns - mg_data%r_gns
|
---|
343 | ! ENDWHERE
|
---|
344 |
|
---|
345 | CALL ropp_io_addvar( &
|
---|
346 | ro_data, &
|
---|
347 | name = "gns_residual_1", &
|
---|
348 | long_name = "1st residual (data-fit) in GNS location", &
|
---|
349 | units = "m", &
|
---|
350 | range = (/-1.0_wp, 1.0_wp/), &
|
---|
351 | DATA = ro_data%Lev1a%r_gns(1:ro_data%Lev1a%npoints, :) - mg_data%r_gns(1:ro_data%Lev1a%npoints, :))
|
---|
352 |
|
---|
353 | ! WHERE (ropp_io_isnan(ro_data%Lev1a%r_leo) .OR. ropp_io_isnan(mg_data%r_leo))
|
---|
354 | ! mg_data%r_leo = ropp_MDFV
|
---|
355 | ! ELSEWHERE
|
---|
356 | ! mg_data%r_leo = ro_data%Lev1a%r_leo - mg_data%r_leo
|
---|
357 | ! ENDWHERE
|
---|
358 |
|
---|
359 | CALL ropp_io_addvar( &
|
---|
360 | ro_data, &
|
---|
361 | name = "leo_residual_1", &
|
---|
362 | long_name = "1st residual (data-fit) in LEO location", &
|
---|
363 | units = "m", &
|
---|
364 | range = (/-1.0_wp, 1.0_wp/), &
|
---|
365 | DATA = ro_data%Lev1a%r_leo(1:ro_data%Lev1a%npoints, :) - mg_data%r_leo(1:ro_data%Lev1a%npoints, :))
|
---|
366 |
|
---|
367 |
|
---|
368 | ! Perform regression on residual positions to gain higher accuracy
|
---|
369 | DO j=1,3
|
---|
370 | CALL ropp_pp_residual_regression(KV,t_norm, &
|
---|
371 | ro_data%Lev1a%r_leo(:,j),coeff_vleo(:,j))
|
---|
372 | ! ro_data%Lev1a%r_leo(:n_cl+n_rs,j),coeff_vleo(:,j))
|
---|
373 | CALL ropp_pp_residual_regression(KV,t_norm, &
|
---|
374 | ro_data%Lev1a%r_gns(:,j),coeff_vgns(:,j))
|
---|
375 | ! ro_data%Lev1a%r_gns(:n_cl+n_rs,j),coeff_vgns(:,j))
|
---|
376 | ENDDO
|
---|
377 | DEALLOCATE(t_norm)
|
---|
378 | DEALLOCATE(KV)
|
---|
379 |
|
---|
380 | DO i=1,ro_data%Lev1a%npoints
|
---|
381 | CALL ropp_pp_interpolate_trajectory(ro_data%Lev1a%dtime, &
|
---|
382 | coeff_vleo, coeff_vgns, &
|
---|
383 | ro_data%georef%r_coc, &
|
---|
384 | ro_data%Lev1a%dtime(i), &
|
---|
385 | mg_data%r_leo(i,:), &
|
---|
386 | mg_data%v_leo(i,:), &
|
---|
387 | mg_data%r_gns(i,:), &
|
---|
388 | mg_data%v_gns(i,:))
|
---|
389 | ENDDO
|
---|
390 |
|
---|
391 | ! WHERE (ropp_io_isnan(ro_data%Lev1a%r_gns(1:ro_data%Lev1a%npoints, :)) .OR. ropp_io_isnan(mg_data%r_gns(1:ro_data%Lev1a%npoints, :)))
|
---|
392 | ! mg_data%r_gns = ropp_MDFV
|
---|
393 | ! ELSEWHERE
|
---|
394 | ! mg_data%r_gns = ro_data%Lev1a%r_gns - mg_data%r_gns
|
---|
395 | ! ENDWHERE
|
---|
396 |
|
---|
397 | CALL ropp_io_addvar( &
|
---|
398 | ro_data, &
|
---|
399 | name = "gns_residual_2", &
|
---|
400 | long_name = "2nd residual (data-fit) in GNS location", &
|
---|
401 | units = "m", &
|
---|
402 | range = (/-1.0_wp, 1.0_wp/), &
|
---|
403 | DATA = ro_data%Lev1a%r_gns(1:ro_data%Lev1a%npoints, :) - mg_data%r_gns(1:ro_data%Lev1a%npoints, :))
|
---|
404 |
|
---|
405 | ! WHERE (ropp_io_isnan(ro_data%Lev1a%r_leo) .OR. ropp_io_isnan(mg_data%r_leo))
|
---|
406 | ! mg_data%r_leo = ropp_MDFV
|
---|
407 | ! ELSEWHERE
|
---|
408 | ! mg_data%r_leo = ro_data%Lev1a%r_leo - mg_data%r_leo
|
---|
409 | ! ENDWHERE
|
---|
410 |
|
---|
411 | CALL ropp_io_addvar( &
|
---|
412 | ro_data, &
|
---|
413 | name = "leo_residual_2", &
|
---|
414 | long_name = "2nd residual (data-fit) in LEO location", &
|
---|
415 | units = "m", &
|
---|
416 | range = (/-1.0_wp, 1.0_wp/), &
|
---|
417 | DATA = ro_data%Lev1a%r_leo(1:ro_data%Lev1a%npoints, :) - mg_data%r_leo(1:ro_data%Lev1a%npoints, :))
|
---|
418 |
|
---|
419 | print*,'ro_data%vlist%vlistd2d%name = ', ro_data%vlist%vlistd2d%name
|
---|
420 | print*,'ro_data%vlist%vlistd2d%next%name = ', ro_data%vlist%vlistd2d%next%name
|
---|
421 | print*,'ro_data%vlist%vlistd2d%next%next%name = ', ro_data%vlist%vlistd2d%next%next%name
|
---|
422 | print*,'ro_data%vlist%vlistd2d%next%next%next%name = ', ro_data%vlist%vlistd2d%next%next%next%name
|
---|
423 | print*,'ro_data%vlist%vlistd2d%next%next%next%next%name = ', ro_data%vlist%vlistd2d%next%next%next%next%name
|
---|
424 |
|
---|
425 | DO i=1,n
|
---|
426 | CALL ropp_pp_interpolate_trajectory(ro_data%Lev1a%dtime, &
|
---|
427 | ! CALL ropp_pp_interpolate_trajectory(ro_data%Lev1a%dtime(:n_cl+n_rs), &
|
---|
428 | coeff_vleo, coeff_vgns, &
|
---|
429 | ro_data%georef%r_coc, &
|
---|
430 | mg_data%dtime(i), &
|
---|
431 | mg_data%r_leo(i,:), &
|
---|
432 | mg_data%v_leo(i,:), &
|
---|
433 | mg_data%r_gns(i,:), &
|
---|
434 | mg_data%v_gns(i,:))
|
---|
435 | ENDDO
|
---|
436 |
|
---|
437 | ! 6.2 Interpolate LCF, phase and amplitude
|
---|
438 |
|
---|
439 | DO i=1,n
|
---|
440 |
|
---|
441 | IF (mrg_Mode) THEN ! Merge LC + RS data
|
---|
442 |
|
---|
443 | i_int = ropp_pp_seek_index(rs_data%dtime, mg_data%dtime(i))
|
---|
444 |
|
---|
445 | IF (i_int == 0 .OR. i_int == rs_data%npoints) THEN
|
---|
446 |
|
---|
447 | ! RS data range outside current time period
|
---|
448 | CALL ropp_pp_interpol(cl_data%dtime, mg_data%dtime(i), &
|
---|
449 | cl_LCF, mg_LCF(i))
|
---|
450 | CALL ropp_pp_interpol(cl_data%dtime, mg_data%dtime(i), &
|
---|
451 | cl_data%snr_L1ca, mg_data%snr_L1ca(i))
|
---|
452 | CALL ropp_pp_interpol(cl_data%dtime, mg_data%dtime(i), &
|
---|
453 | cl_data%snr_L1p, mg_data%snr_L1p(i))
|
---|
454 | CALL ropp_pp_interpol(cl_data%dtime, mg_data%dtime(i), &
|
---|
455 | cl_data%snr_L2p, mg_data%snr_L2p(i))
|
---|
456 | CALL ropp_pp_interpol(cl_data%dtime, mg_data%dtime(i), &
|
---|
457 | cl_data%phase_L1, mg_data%phase_L1(i))
|
---|
458 | CALL ropp_pp_interpol(cl_data%dtime, mg_data%dtime(i), &
|
---|
459 | cl_data%phase_L2, mg_data%phase_L2(i))
|
---|
460 |
|
---|
461 | ELSE
|
---|
462 |
|
---|
463 | ! RS data range within current time period
|
---|
464 | CALL ropp_pp_interpol(rs_data%dtime, mg_data%dtime(i), &
|
---|
465 | rs_LCF, mg_LCF(i))
|
---|
466 | CALL ropp_pp_interpol(rs_data%dtime, mg_data%dtime(i), &
|
---|
467 | rs_data%snr_L1ca, mg_data%snr_L1ca(i))
|
---|
468 | CALL ropp_pp_interpol(rs_data%dtime, mg_data%dtime(i), &
|
---|
469 | rs_data%snr_L1p, mg_data%snr_L1p(i))
|
---|
470 | CALL ropp_pp_interpol(rs_data%dtime, mg_data%dtime(i), &
|
---|
471 | rs_data%snr_L2p, mg_data%snr_L2p(i))
|
---|
472 | CALL ropp_pp_interpol(rs_data%dtime, mg_data%dtime(i), &
|
---|
473 | rs_data%phase_L1, mg_data%phase_L1(i))
|
---|
474 | CALL ropp_pp_interpol(rs_data%dtime, mg_data%dtime(i), &
|
---|
475 | rs_data%phase_L2, mg_data%phase_L2(i))
|
---|
476 |
|
---|
477 | ENDIF
|
---|
478 |
|
---|
479 | ELSE ! Use only CL data
|
---|
480 |
|
---|
481 | CALL ropp_pp_interpol(cl_data%dtime, mg_data%dtime(i), &
|
---|
482 | cl_LCF, mg_LCF(i))
|
---|
483 | CALL ropp_pp_interpol(cl_data%dtime, mg_data%dtime(i), &
|
---|
484 | cl_data%snr_L1ca, mg_data%snr_L1ca(i))
|
---|
485 | CALL ropp_pp_interpol(cl_data%dtime, mg_data%dtime(i), &
|
---|
486 | cl_data%snr_L1p, mg_data%snr_L1p(i))
|
---|
487 | CALL ropp_pp_interpol(cl_data%dtime, mg_data%dtime(i), &
|
---|
488 | cl_data%snr_L2p, mg_data%snr_L2p(i))
|
---|
489 | CALL ropp_pp_interpol(cl_data%dtime, mg_data%dtime(i), &
|
---|
490 | cl_data%phase_L1, mg_data%phase_L1(i))
|
---|
491 | CALL ropp_pp_interpol(cl_data%dtime, mg_data%dtime(i), &
|
---|
492 | cl_data%phase_L2, mg_data%phase_L2(i))
|
---|
493 |
|
---|
494 | ENDIF
|
---|
495 |
|
---|
496 |
|
---|
497 | ENDDO
|
---|
498 |
|
---|
499 | !-------------------------------------------------------------------------------
|
---|
500 | ! 7. Connect phase
|
---|
501 | !-------------------------------------------------------------------------------
|
---|
502 |
|
---|
503 | ! 7.1 Calculate MSIS excess phase on merged time grid
|
---|
504 | CALL occ_point( mg_data%r_leo, mg_data%r_gns, &
|
---|
505 | ro_data%georef%lat, ro_data%georef%lon, &
|
---|
506 | ro_data%georef%r_coc, ro_data%georef%roc, &
|
---|
507 | ro_data%georef%azimuth, &
|
---|
508 | ro_data%georef%undulation, &
|
---|
509 | config%egm96, config%corr_egm96)
|
---|
510 |
|
---|
511 | ALLOCATE(mg_msis(n))
|
---|
512 | ALLOCATE(mg_impact(n))
|
---|
513 | CALL ropp_pp_modelphase(ro_data%dtocc%month, ro_data%georef%lat, &
|
---|
514 | ro_data%georef%lon, mg_data%dtime, &
|
---|
515 | mg_data%r_leo, mg_data%r_gns, &
|
---|
516 | ro_data%georef%r_coc, ro_data%georef%roc, &
|
---|
517 | mg_msis, mg_impact, config)
|
---|
518 | DEALLOCATE(mg_impact)
|
---|
519 |
|
---|
520 | ! 7.2 Phase re-accumulation
|
---|
521 |
|
---|
522 | ALLOCATE(mg_ds(n))
|
---|
523 | WHERE (.NOT. BTEST(mg_LCF(:),3))
|
---|
524 | mg_ds(:) = (2.0_wp*Pi*f_L1/c_light)*(mg_data%phase_L1(:)-mg_msis(:))
|
---|
525 | ELSEWHERE
|
---|
526 | mg_ds(:) = 0.0_wp
|
---|
527 | ENDWHERE
|
---|
528 |
|
---|
529 | CALL Accumulate_Phase(mg_DS)
|
---|
530 |
|
---|
531 | !-------------------------------------------------------------------------------
|
---|
532 | ! 8. Fill in data gaps
|
---|
533 | !-------------------------------------------------------------------------------
|
---|
534 |
|
---|
535 | ts = MINVAL(ABS(cl_data%dtime(2:n_cl) - cl_data%dtime(1:n_cl-1)))
|
---|
536 |
|
---|
537 | ig1 = 1
|
---|
538 |
|
---|
539 | DO i=2,n
|
---|
540 |
|
---|
541 | IF (BTEST(mg_LCF(i),3) .AND. .NOT. BTEST(mg_LCF(i-1),3)) THEN
|
---|
542 | ig1 = i-1
|
---|
543 | ENDIF
|
---|
544 |
|
---|
545 | IF (.NOT. BTEST(mg_LCF(i),3) .AND. BTEST(mg_LCF(i-1),3)) THEN
|
---|
546 |
|
---|
547 | ig2 = i
|
---|
548 | igl = ig2-ig1
|
---|
549 | tgl = mg_data%dtime(ig2) - mg_data%dtime(ig1)
|
---|
550 |
|
---|
551 | IF (tgl < gap) THEN ! Fill in data gap
|
---|
552 |
|
---|
553 | DO j=ig1+1,ig2-1
|
---|
554 | CALL ropp_pp_interpol(mg_data%dtime(ig1:ig2:igl), mg_data%dtime(j), &
|
---|
555 | mg_ds(ig1:ig2:igl), mg_ds(j))
|
---|
556 | mg_LCF(j) = IBCLR(mg_LCF(j),3)
|
---|
557 | ENDDO
|
---|
558 | ENDIF
|
---|
559 |
|
---|
560 | ENDIF
|
---|
561 |
|
---|
562 | ENDDO
|
---|
563 |
|
---|
564 | !-------------------------------------------------------------------------------
|
---|
565 | ! 9. Restore and normalise phase
|
---|
566 | !-------------------------------------------------------------------------------
|
---|
567 |
|
---|
568 | ! 9.1 Restore phase variation
|
---|
569 |
|
---|
570 | mg_data%phase_L1(:) = mg_msis(:) + (c_light/(2.0_wp*pi*f_L1))*mg_ds(:)
|
---|
571 |
|
---|
572 | ! 9.2 Normalise to zero minimum value
|
---|
573 |
|
---|
574 | sb = MINVAL(mg_data%phase_L1(:))
|
---|
575 | mg_data%phase_L1(:) = mg_data%phase_L1(:) - sb
|
---|
576 |
|
---|
577 | !-------------------------------------------------------------------------------
|
---|
578 | ! 7. Setup merged output data
|
---|
579 | !-------------------------------------------------------------------------------
|
---|
580 |
|
---|
581 | CALL ropp_io_free(ro_data%Lev1a)
|
---|
582 | DEALLOCATE(LCF)
|
---|
583 |
|
---|
584 | CALL ropp_io_init(ro_data%Lev1a, n)
|
---|
585 | ALLOCATE(LCF(n))
|
---|
586 |
|
---|
587 | ro_data%Lev1a = mg_data
|
---|
588 | LCF(:) = mg_LCF(:)
|
---|
589 |
|
---|
590 | !-------------------------------------------------------------------------------
|
---|
591 | ! 8. Re-compute occultation point
|
---|
592 | !-------------------------------------------------------------------------------
|
---|
593 |
|
---|
594 | CALL occ_point( ro_data%Lev1a%r_leo, ro_data%Lev1a%r_gns, &
|
---|
595 | ro_data%georef%lat, ro_data%georef%lon, &
|
---|
596 | ro_data%georef%r_coc, ro_data%georef%roc, &
|
---|
597 | ro_data%georef%azimuth, &
|
---|
598 | ro_data%georef%undulation, &
|
---|
599 | config%egm96, config%corr_egm96)
|
---|
600 |
|
---|
601 | !-------------------------------------------------------------------------------
|
---|
602 | ! 9. Quality control
|
---|
603 | !-------------------------------------------------------------------------------
|
---|
604 |
|
---|
605 | WHERE(ro_data%Lev1a%phase_L1(:) == ropp_MDFV)
|
---|
606 | ro_data%Lev1a%phase_L1(:) = -1.0_wp
|
---|
607 | ENDWHERE
|
---|
608 | WHERE(ro_data%Lev1a%phase_L2(:) == ropp_MDFV)
|
---|
609 | ro_data%Lev1a%phase_L2(:) = -1.0_wp
|
---|
610 | ENDWHERE
|
---|
611 |
|
---|
612 | !-------------------------------------------------------------------------------
|
---|
613 | ! 10. Clean up
|
---|
614 | !-------------------------------------------------------------------------------
|
---|
615 |
|
---|
616 | DEALLOCATE(mg_msis)
|
---|
617 | CALL ropp_io_free(cl_data)
|
---|
618 | CALL ropp_io_free(rs_data)
|
---|
619 |
|
---|
620 | print*,'ro_data%vlist%vlistd2d%name = ', ro_data%vlist%vlistd2d%name
|
---|
621 | print*,'ro_data%vlist%vlistd2d%data(1:5, 1:2) = ', ro_data%vlist%vlistd2d%data(1:5, 1:2)
|
---|
622 |
|
---|
623 | print*,'ro_data%vlist%vlistd2d%next%name = ', ro_data%vlist%vlistd2d%next%name
|
---|
624 | print*,'ro_data%vlist%vlistd2d%next%data(1:5, 1:3) = ', ro_data%vlist%vlistd2d%next%data(1:5, 1:3)
|
---|
625 |
|
---|
626 | print*,'ro_data%vlist%vlistd2d%next%next%name = ', ro_data%vlist%vlistd2d%next%next%name
|
---|
627 | print*,'ro_data%vlist%vlistd2d%next%next%data(1:5, 1:3) = ', ro_data%vlist%vlistd2d%next%next%data(1:5, 1:3)
|
---|
628 |
|
---|
629 | print*,'ro_data%vlist%vlistd2d%next%next%next%name = ', ro_data%vlist%vlistd2d%next%next%next%name
|
---|
630 | print*,'ro_data%vlist%vlistd2d%next%next%next%data(1:5, 1:3) = ', ro_data%vlist%vlistd2d%next%next%next%data(1:5, 1:3)
|
---|
631 |
|
---|
632 | print*,'ro_data%vlist%vlistd2d%next%next%next%next%name = ', ro_data%vlist%vlistd2d%next%next%next%next%name
|
---|
633 | print*,'ro_data%vlist%vlistd2d%next%next%next%next%data(1:5, 1:3) = ', ro_data%vlist%vlistd2d%next%next%next%next%data(1:5, 1:3)
|
---|
634 |
|
---|
635 | IF(ASSOCIATED(ro_data%vlist%VlistD0d)) print*,'SHAPE(ro_data%vlist%VlistD0d) = ', SHAPE(ro_data%vlist%VlistD0d)
|
---|
636 | IF(ASSOCIATED(ro_data%vlist%VlistD1d)) print*,'SHAPE(ro_data%vlist%VlistD1d) = ', SHAPE(ro_data%vlist%VlistD1d)
|
---|
637 | IF(ASSOCIATED(ro_data%vlist%VlistD2d)) print*,'SHAPE(ro_data%vlist%VlistD2d) = ', SHAPE(ro_data%vlist%VlistD2d)
|
---|
638 |
|
---|
639 | CONTAINS
|
---|
640 |
|
---|
641 | !-------------------------------------------------------------------------------
|
---|
642 | ! 11. Transform phase to accumulated phase
|
---|
643 | !-------------------------------------------------------------------------------
|
---|
644 |
|
---|
645 | SUBROUTINE Accumulate_Phase(Ph, Sign) ! (Array of (accumulated) phase, dir)
|
---|
646 |
|
---|
647 | ! Method:
|
---|
648 | ! Sign = 0 or no Sign:
|
---|
649 | ! Adding +-2*Pi where phase jumps from
|
---|
650 | ! +-Pi to -+Pi,
|
---|
651 | ! Sign > 0:
|
---|
652 | ! Adding +2*Pi where phase jumps from
|
---|
653 | ! - to +
|
---|
654 | ! Sign < 0
|
---|
655 | ! Adding -2*Pi where phase jumps from
|
---|
656 | ! + to -
|
---|
657 |
|
---|
658 | ! 11.1 Declarations
|
---|
659 |
|
---|
660 | USE typesizes, ONLY: wp => EightByteReal
|
---|
661 | USE ropp_pp_constants, ONLY: pi
|
---|
662 | IMPLICIT NONE
|
---|
663 |
|
---|
664 | REAL(wp), DIMENSION(:), INTENT(inout) :: Ph ! Phase --> accumulated phase
|
---|
665 | INTEGER, OPTIONAL, INTENT(in) :: Sign ! Phase change sign
|
---|
666 |
|
---|
667 | INTEGER :: i ! Array index
|
---|
668 | INTEGER :: PSign ! Phase change sign
|
---|
669 |
|
---|
670 | ! 11.2 Determine phase change sign
|
---|
671 |
|
---|
672 | IF (.NOT. PRESENT(Sign)) THEN
|
---|
673 | PSign = 0
|
---|
674 | ELSE
|
---|
675 | PSign = Sign
|
---|
676 | ENDIF
|
---|
677 |
|
---|
678 | ! 11.3 Accumulate phase
|
---|
679 |
|
---|
680 | IF (PSign == 0) THEN
|
---|
681 | DO i=2,SIZE(Ph)
|
---|
682 | Ph(i) = Ph(i-1) + MODULO(Ph(i)-Ph(i-1)+pi, 2*pi) - pi
|
---|
683 | ENDDO
|
---|
684 | ELSEIF (PSign > 0) THEN
|
---|
685 | DO i=2,SIZE(Ph)
|
---|
686 | Ph(i) = Ph(i-1) + MODULO(Ph(i)-Ph(i-1), 2*pi)
|
---|
687 | ENDDO
|
---|
688 | ELSEIF (PSign < 0) THEN
|
---|
689 | DO i=2,SIZE(Ph)
|
---|
690 | Ph(i) = Ph(i-1) + MODULO(Ph(i)-Ph(i-1)+2*pi, 2*pi) - 2*pi
|
---|
691 | ENDDO
|
---|
692 | ENDIF
|
---|
693 |
|
---|
694 | END SUBROUTINE Accumulate_Phase
|
---|
695 |
|
---|
696 |
|
---|
697 |
|
---|
698 |
|
---|
699 | FUNCTION ropp_io_isnan(x) RESULT(lnan) ! Says where reals are NaNs.
|
---|
700 |
|
---|
701 | USE typesizes, ONLY: wp => EightByteReal
|
---|
702 |
|
---|
703 | IMPLICIT NONE
|
---|
704 |
|
---|
705 | INTEGER :: IPInf, IMinf
|
---|
706 | REAL(wp) :: PInf, MInf
|
---|
707 | data IPInf/B'01111111100000000000000000000000'/ ! +Infinity
|
---|
708 | data IMInf/B'11111111100000000000000000000000'/ ! -Infinity
|
---|
709 |
|
---|
710 | REAL(wp), DIMENSION(:,:), INTENT(IN) :: x
|
---|
711 | LOGICAL, DIMENSION(SIZE(x,1), SIZE(x,2)) :: lnan
|
---|
712 |
|
---|
713 | PInf = TRANSFER(IPinf, Pinf)
|
---|
714 | Minf = TRANSFER(IMinf, Minf)
|
---|
715 |
|
---|
716 | lnan = .FALSE.
|
---|
717 |
|
---|
718 | ! WHERE ((x /= x) .OR. (x + 1.0_wp == x) .OR. ((x > 0) .EQV. (x <= 0)) .OR. (x == Pinf) .OR. (x == Minf)) &
|
---|
719 | WHERE ((x /= x) .OR. (x + 1.0_wp == x) .OR. ((x > 0) .EQV. (x <= 0))) &
|
---|
720 | lnan = .TRUE.
|
---|
721 |
|
---|
722 | END FUNCTION ropp_io_isnan
|
---|
723 |
|
---|
724 | END SUBROUTINE ropp_pp_preprocess_GRASRS
|
---|