1

I want to minimize the spikes caused by cloud cover (smooth) NDVI timeseries data before applying Savitzky–Golay filter in Timesat. My previous post was also related to timeseries (Interpreting NDVI curve in Timesat?). I have gone through a paper "Extraction of Agricultural Phenological Parameters of Sri Lanka Using MODIS, NDVI Time Series Data(DOI: 10.1016/j.profoo.2016.02.027) The author used periodicity converter tool in SPIRITS to minimize the spikes. I am also using similar tool "processing-> temporal-> smoothing" and also used periodicity converter tool but I ended up with error(below)

-------------------------------------------------------------------------------------------------------------------------------

2018/11/13 05:34:01 STATE: RUNNING 2018/11/13 05:34:01 INFO: Task SMOOTH15_ISA.exe 2018/11/13 05:34:01 INFO: parameter(0) : 'GUI' 2018/11/13 05:34:01 INFO: parameter(1) : '10' 2018/11/13 05:34:01 INFO: parameter(2) : '1' 2018/11/13 05:34:01 INFO: parameter(3) : '36' 2018/11/13 05:34:01 INFO: parameter(4) : '20150101' 2018/11/13 05:34:01 INFO: parameter(5) : '20171231' 2018/11/13 05:34:01 INFO: parameter(6) : 'G:\Spirits\3yrs\Image_File\vt' 2018/11/13 05:34:01 INFO: parameter(7) : '6' 2018/11/13 05:34:01 INFO: parameter(8) : 'i' 2018/11/13 05:34:01 INFO: parameter(9) : '0' 2018/11/13 05:34:01 INFO: parameter(10) : '0' 2018/11/13 05:34:01 INFO: parameter(11) : ' ' 2018/11/13 05:34:01 INFO: parameter(12) : ' ' 2018/11/13 05:34:01 INFO: parameter(13) : ' ' 2018/11/13 05:34:01 INFO: parameter(14) : '0' 2018/11/13 05:34:01 INFO: parameter(15) : '0' 2018/11/13 05:34:01 INFO: parameter(16) : '10' 2018/11/13 05:34:01 INFO: parameter(17) : '1' 2018/11/13 05:34:01 INFO: parameter(18) : '36' 2018/11/13 05:34:01 INFO: parameter(19) : '20150101' 2018/11/13 05:34:01 INFO: parameter(20) : '20171231' 2018/11/13 05:34:01 INFO: parameter(21) : 'G:\output\smooth\sw_vt' 2018/11/13 05:34:01 INFO: parameter(22) : '1' 2018/11/13 05:34:01 INFO: parameter(23) : 'i' 2018/11/13 05:34:01 INFO: parameter(24) : ' ' 2018/11/13 05:34:01 INFO: parameter(25) : ' ' 2018/11/13 05:34:01 INFO: parameter(26) : ' 0.0' 2018/11/13 05:34:01 INFO: parameter(27) : '75' 2018/11/13 05:34:01 INFO: parameter(28) : '0' 2018/11/13 05:34:01 INFO: parameter(29) : '1' 2018/11/13 05:34:01 INFO: parameter(30) : '2' 2018/11/13 05:34:01 INFO: parameter(31) : 'G:\Downloads\SpiritsExtract_152(1)\libs\GLIMPSE\GLIMPSE_TMP\sm14sm_6561240588128570923' 2018/11/13 05:34:01 INFO:


SMOOTH15: SMOOTH SERIES of any VI (NDVI, fAPAR,...) & MODIFY the PERIODICITY


IN-SERIES: Periodicity                                                     

p1.Periodicity in Days [1-30]. NB: 10=def=civil dekads, 30=civil months
p2.Account for Leap Years [0/1=def]. If 0, the series skips 29 February.
p3.Periods per Year. Skip if period=1/5/10 or 30 (if p1=5 => 73 periods/year) IN-SERIES: Start/End of entire series of IN-IMGs (incl. PREV/HIST, p10-13) p4.IN-date1 [YYYYMMDD] NB: - IN-Dates (p4/5) are reset to start of period
p5.IN-date2 [YYYYMMDD] - Period p4-p5 must at least cover 180 days
IN-SERIES: NAMES OF ACTUAL IMGs: PiDiSi.img (BYTE, fixed MAPINFO/VALUES!)
p6.Pi=Prefix (add drive/path if needed)
p7.Di=Date-Format [1-6] (6 different formats, see below)
p8.Si=Suffix (no extension)
1.YYYYMMDD 4.YYmDD Year : YYYY[1950-2049] or YY[50-49]
2.YYMMDD 5.YYYYPP Month: MM[01-12] or m[A=Jan...L=Dec]
3.YYYYmDD 6.YYPP Day: DD[01-31] Period_in_year: PP[(0)01-(n)nn]
IN-SERIES: OTHER PARAMETERS concerning the TIME SERIES
p9.Max. Nr. consecutively missing ACT IMGs [def=0, max=90] p10.IMGs of Pre/Hist Year replace missing IMGs at edges [def=0=No,1=Pre,2=Hist] IN-SERIES: NAMES OF these IMGs of PREV/HIST YEAR: PhDhSh.img
p11.Ph=Prefix (add drive/path if needed - If p10=0 => Skip p11-p13) p12.Dh=Date-Format [1-6] (6 formats, see above - HIST IMGs have YYYY=1962!)
p13.Sh=Suffix (no extension)
IN-SERIES: PROFILE TAILS: Improve extrapolation at start/end of IN-series
NB: Only allowed if IN-series covers exactly one/more full years, else Skip p14.Periods to copy from Profile-End before Profile-Start [def=0=none] p15. from Profile-Start after Profile-End [def=0=none] OUT-SERIES: Periodicity
p16.Periodicity in Days [1-30]. NB: 10=def=civil dekads, 30=civil months
p17.Account for Leap Years [0/1=def]. If 0, the series skips 29 February.
p18.Periods per Year. Skip if period=1/5/10 or 30 (if p18=5 => 73 periods/year) OUT-SERIES: Start/End dates
p19.OUT-Date1 [YYYYMMDD] NB: - OUT-Dates (p19/20) are reset to start of period p20.OUT-Date2 [YYYYMMDD] - Defaults: adapted from IN-Dates (p4-p5)
OUT-SERIES: NAMES OF OUT-IMGs: PoDoSo.img
p21.Po=Prefix (add drive/path if needed)
p22.Do=Date-Format [1-6] (6 diff. formats - see above)
p23.So=Suffix (no extension)
OTHER PARAMETERS concerning the TIME SERIES OF IN/OUT-IMGs
p24.Name (no ext.) of VAR/MTA-files with names of OUT-IMGs (blank=none)
p25.Lowest physical value Ymin for cloudfree land pixels [def=Vint+Vslo*Vlo] p26.Eliminate local Minima if Y-difference to neighbours > YmaxDip [def=0=skip] p27.Max. % of missing values in each pixel-profile [0-99, def=75] p28.Extrapolate 'TAILS' with missing IN-values [def=0/1] 0=>Normal pixels may have flagged TAILS at start/end of profile
1=>All normal pixels will only contain 'good' (non-flagged) OUT-values
p29.Method for OUT-Flags (most often only in profile TAILS) [def=0/1/2] Always possible: 0 = Only 2 flags (254=water, 251=missing over land) IN/OUT-IMGs UNIflagged: 1 = Minimize OUT-flagged observations
(251-255) 2 = Recover IN-flags 251=Missing and 253=Snow
p30.Smoothing: 0=MEAN, 1=BISE (only S10-inputs), 2=def=SWETS, 3=WHITTAKER
p31.SPP-file with fixed Specifications (ASCII, no ext. *.SPP)
OPTIONAL OUT-IMGs with QUALITY INFORMATION and NAMES PqDqSq.img
p32.Pq=Prefix (add drive/path if needed - blank = none)
p33.Dq=Date-Format [1-6] (6 diff. formats - see above)
p34.Sq=Suffix (no extension)
p35.Max. allowed difference Ydelt: Smoothed <> PRE-treated IN-values [def=0.0]


IN-SERIES: Periodicity

p1.Periodicity in Days [1-30, def=10] =>10 p2.Account for Leap Years [0/1=def] =>1 p3.Periods per Year [fixed] =>36 IN-PERIOD: Start/End of complete series of IN-IMGs (incl. PREV/HIST) p4.Date in FIRST IN-Period [YYYYMMDD] =>20150101 p5.Date in LAST IN-period [YYYYMMDD] =>20171231 Nr. of IN-Periods : 108 (or circa 1080 days) NAMES OF ACTUAL/REAL IN-IMGs: PiDiSi.img p6.Prefix Pi (add drive/path if needed) =>G:\Spirits\3yrs\Image_File\vt 1.YYYYMMDD 4.YYmDD Year : YYYY[1950-2049] or YY[50-49]
2.YYMMDD 5.YYYYPP Month: MM[01-12] or m[A=Jan...L=Dec] 3.YYYYmDD 6.YYPP Day: DD[01-31] Period_in_year: PP[01-36] p7.Date-Format Di (see above) [1-6] =>6 p8.Suffix Si (no extension: .IMG) =>i OTHER PARAMETERS concerning the TIME SERIES OF IN-IMGs p9.Max. Nr. of consec. missing ACT. IMGs in centre of series =>0 p10.Extend Series at edges: 0=No, 1=Prev.IMGs, 2=Hist.IMGs =>0 Nr. of ACTUAL IN-IMGs found: 0 ( 0.00% of total periods 108) 2018/11/13 05:34:01 ERROR: SMOOTH15-INPUT STAGE: Insuffient number of ACTUAL IN-IMGs (0) found 2018/11/13 05:34:01 STATE: ERROR 2018/11/13 05:34:01 ERROR: Program ended with exit code: 1

-------------------------------------------------------------------------------------------------------------------------------

NOTE: I have input 3 years NDVI MOD13Q1 images (.img, .hdr format)

ENVI description = { G:\Spirits\3yrs\Image_File\2015_1.img} samples = 8709 lines = 4636 bands = 1 header offset = 0 file type = ENVI Standard data type = 2 interleave = bsq byte order = 0 map info = {Geographic Lat/Lon, 1, 1, 85.13811073, 29.99895833, 0.00215682537558847, 0.00215682537553926,WGS-84} coordinate system string = {GEOGCS["GCS_WGS_1984",DATUM["D_WGS_1984",SPHEROID["WGS_1984",6378137,298.257223563]],PRIMEM["Greenwich",0],UNIT["Degree",0.017453292519943295]]} band names = { Band 1}

I am unable to figure out the error caused. Anyone??

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.