CDO Users Guide

Climate Data Operators

Version 1.4.0
October 2009

Uwe Schulzweida, Luis Kornblueh MPI for Meteorology

Ralf Quast Brockmann Consult

Contents
1. Introduction
1.1. Building from sources . . . .
1.1.1. Compilation . . . . . .
1.1.2. Installation . . . . . .
1.2. Usage . . . . . . . . . . . . .
1.2.1. Options . . . . . . . .
1.2.2. Operators . . . . . . .
1.2.3. Combining operators .
1.2.4. Operator parameter .
1.3. Grid description . . . . . . .
1.3.1. Predefined grids . . .
1.3.2. Grids from data files .
1.3.3. SCRIP grids . . . . .
1.3.4. PINGO grids . . . . .
1.3.5. CDO grids . . . . . .
1.4. Z-axis description . . . . . . .
1.5. Time axis . . . . . . . . . . .
1.5.1. Absolute time . . . . .
1.5.2. Relative time . . . . .
1.5.3. Conversion of the time
1.6. Parameter table . . . . . . . .
1.7. Missing values . . . . . . . . .
1.7.1. Mean and average . .

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

2. Reference manual
2.1. Information . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1.1. INFO - Information and simple statistics . . . . .
2.1.2. SINFO - Short information . . . . . . . . . . . . .
2.1.3. DIFF - Compare two datasets field by field . . . .
2.1.4. NINFO - Print the number of parameters, levels or
2.1.5. SHOWINFO - Show variables, levels or times . . .
2.1.6. FILEDES - Dataset description . . . . . . . . . . .
2.2. File operations . . . . . . . . . . . . . . . . . . . . . . . .
2.2.1. COPY - Copy datasets . . . . . . . . . . . . . . . .
2.2.2. REPLACE - Replace variables . . . . . . . . . . .
2.2.3. MERGE - Merge datasets . . . . . . . . . . . . . .
2.2.4. SPLIT - Split a dataset . . . . . . . . . . . . . . .
2.2.5. SPLITTIME - Split time steps of a dataset . . . .
2.2.6. SPLITSEL - Split selected time steps . . . . . . .
2.3. Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3.1. SELECT - Select fields . . . . . . . . . . . . . . . .
2.3.2. SELTIME - Select time steps . . . . . . . . . . . .
2.3.3. SELBOX - Select a box of a field . . . . . . . . . .
2.4. Conditional selection . . . . . . . . . . . . . . . . . . . . .
2.4.1. COND - Conditional select one field . . . . . . . .
2.4.2. COND2 - Conditional select two fields . . . . . . .
2.4.3. CONDC - Conditional select a constant . . . . . .
2.5. Comparison . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.1. COMP - Comparison of two fields . . . . . . . . .

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

6
6
7
7
8
8
8
8
9
9
10
10
10
11
11
12
13
13
14
14
14
14
15

. . . .
. . . .
. . . .
. . . .
times
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

16
17
18
19
20
21
22
23
24
25
25
26
27
28
29
30
31
33
35
36
37
37
38
39
40

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

Contents
2.5.2. COMPC - Comparison of a field with a constant . . . . . .
2.6. Modification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.6.1. SET - Set field info . . . . . . . . . . . . . . . . . . . . . . .
2.6.2. SETTIME - Set time . . . . . . . . . . . . . . . . . . . . . .
2.6.3. CHANGE - Change field header . . . . . . . . . . . . . . .
2.6.4. SETGRID - Set grid type . . . . . . . . . . . . . . . . . . .
2.6.5. SETZAXIS - Set z-axis type . . . . . . . . . . . . . . . . . .
2.6.6. SETGATT - Set global attribute . . . . . . . . . . . . . . .
2.6.7. INVERT - Invert latitudes . . . . . . . . . . . . . . . . . . .
2.6.8. INVERTLEV - Invert levels . . . . . . . . . . . . . . . . . .
2.6.9. MASKREGION - Mask regions . . . . . . . . . . . . . . . .
2.6.10. MASKBOX - Mask a box . . . . . . . . . . . . . . . . . . .
2.6.11. SETBOX - Set a box to constant . . . . . . . . . . . . . . .
2.6.12. ENLARGE - Enlarge fields . . . . . . . . . . . . . . . . . .
2.6.13. SETMISS - Set missing value . . . . . . . . . . . . . . . . .
2.7. Arithmetic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.7.1. EXPR - Evaluate expressions . . . . . . . . . . . . . . . . .
2.7.2. MATH - Mathematical functions . . . . . . . . . . . . . . .
2.7.3. ARITHC - Arithmetic with a constant . . . . . . . . . . . .
2.7.4. ARITH - Arithmetic on two datasets . . . . . . . . . . . . .
2.7.5. MONARITH - Monthly arithmetic . . . . . . . . . . . . . .
2.7.6. YMONARITH - Multi-year monthly arithmetic . . . . . . .
2.7.7. ARITHDAYS - Arithmetic with days . . . . . . . . . . . .
2.8. Statistical values . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.8.1. ENSSTAT - Statistical values over an ensemble . . . . . . .
2.8.2. FLDSTAT - Statistical values over a field . . . . . . . . . .
2.8.3. ZONSTAT - Zonal statistical values . . . . . . . . . . . . .
2.8.4. MERSTAT - Meridional statistical values . . . . . . . . . .
2.8.5. VERTSTAT - Vertical statistical values . . . . . . . . . . .
2.8.6. TIMSELSTAT - Time range statistical values . . . . . . . .
2.8.7. TIMSELPCTL - Time range percentile values . . . . . . . .
2.8.8. RUNSTAT - Running statistical values . . . . . . . . . . . .
2.8.9. RUNPCTL - Running percentile values . . . . . . . . . . .
2.8.10. TIMSTAT - Statistical values over all time steps . . . . . .
2.8.11. TIMPCTL - Percentile values over all time steps . . . . . .
2.8.12. HOURSTAT - Hourly statistical values . . . . . . . . . . .
2.8.13. HOURPCTL - Hourly percentile values . . . . . . . . . . .
2.8.14. DAYSTAT - Daily statistical values . . . . . . . . . . . . .
2.8.15. DAYPCTL - Daily percentile values . . . . . . . . . . . . .
2.8.16. MONSTAT - Monthly statistical values . . . . . . . . . . .
2.8.17. MONPCTL - Monthly percentile values . . . . . . . . . . .
2.8.18. YEARSTAT - Yearly statistical values . . . . . . . . . . . .
2.8.19. YEARPCTL - Yearly percentile values . . . . . . . . . . . .
2.8.20. SEASSTAT - Seasonal statistical values . . . . . . . . . . .
2.8.21. SEASPCTL - Seasonal percentile values . . . . . . . . . . .
2.8.22. YHOURSTAT - Multi-year hourly statistical values . . . .
2.8.23. YDAYSTAT - Multi-year daily statistical values . . . . . .
2.8.24. YDAYPCTL - Multi-year daily percentile values . . . . . .
2.8.25. YMONSTAT - Multi-year monthly statistical values . . . .
2.8.26. YMONPCTL - Multi-year monthly percentile values . . . .
2.8.27. YSEASSTAT - Multi-year seasonal statistical values . . . .
2.8.28. YSEASPCTL - Multi-year seasonal percentile values . . . .
2.8.29. YDRUNSTAT - Multi-year daily running statistical values .
2.8.30. YDRUNPCTL - Multi-year daily running percentile values
2.9. Regression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.9.1. REGRES - Regression . . . . . . . . . . . . . . . . . . . . .
2.9.2. DETREND - Detrend time series . . . . . . . . . . . . . . .
2.9.3. TREND - Trend of time series . . . . . . . . . . . . . . . .

Contents
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

41
42
43
44
46
47
47
48
49
49
50
51
52
53
54
56
57
58
59
60
61
62
63
64
68
70
72
74
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
102
103
104
104
105

Contents
2.9.4. SUBTREND - Subtract a trend . . . . . . . . . . . . . . . . . . . . . . . . .
2.10. Interpolation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.10.1. REMAPGRID - SCRIP grid interpolation . . . . . . . . . . . . . . . . . . .
2.10.2. GENWEIGHTS - Generate SCRIP grid interpolation weights . . . . . . . .
2.10.3. REMAP - SCRIP grid remapping . . . . . . . . . . . . . . . . . . . . . . .
2.10.4. INTGRID - Grid interpolation . . . . . . . . . . . . . . . . . . . . . . . . .
2.10.5. REMAPETA - Remap vertical hybrid level . . . . . . . . . . . . . . . . . .
2.10.6. INTVERT - Vertical interpolation . . . . . . . . . . . . . . . . . . . . . . .
2.10.7. INTLEVEL - Linear level interpolation . . . . . . . . . . . . . . . . . . . .
2.10.8. INTTIME - Time interpolation . . . . . . . . . . . . . . . . . . . . . . . . .
2.10.9. INTYEAR - Year interpolation . . . . . . . . . . . . . . . . . . . . . . . . .
2.11. Transformation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.11.1. SPECTRAL - Spectral transformation . . . . . . . . . . . . . . . . . . . . .
2.11.2. WIND - Wind transformation . . . . . . . . . . . . . . . . . . . . . . . . . .
2.12. Formatted I/O . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.12.1. INPUT - Formatted input . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.12.2. OUTPUT - Formatted output . . . . . . . . . . . . . . . . . . . . . . . . .
2.13. Miscellaneous . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.13.1. GRIDCELL - Grid cell quantities . . . . . . . . . . . . . . . . . . . . . . . .
2.13.2. GRADSDES - GrADS data descriptor file . . . . . . . . . . . . . . . . . . .
2.13.3. SMOOTH9 - 9 point smoothing . . . . . . . . . . . . . . . . . . . . . . . . .
2.13.4. SETRANGE - Set range to constant . . . . . . . . . . . . . . . . . . . . . .
2.13.5. TIMSORT - Timsort . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.13.6. VARGEN - Generate a field . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.13.7. ROTUV - Rotation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.13.8. MASTRFU - Mass stream function . . . . . . . . . . . . . . . . . . . . . . .
2.13.9. HISTOGRAM - Histogram . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.13.10.WCT - Windchill temperature . . . . . . . . . . . . . . . . . . . . . . . . .
2.13.11.FDNS - Frost days where no snow index per time period . . . . . . . . . . .
2.13.12.STRWIN - Strong wind days index per time period . . . . . . . . . . . . . .
2.13.13.STRBRE - Strong breeze days index per time period . . . . . . . . . . . . .
2.13.14.STRGAL - Strong gale days index per time period . . . . . . . . . . . . . .
2.13.15.HURR - Hurricane days index per time period . . . . . . . . . . . . . . . .
2.13.16.IMPORTAMSR - Import AMSR binary files . . . . . . . . . . . . . . . . .
2.13.17.IMPORTCMSAF - Import CM-SAF HDF5 files . . . . . . . . . . . . . . .
2.13.18.IMPORTBINARY - Import binary data sets . . . . . . . . . . . . . . . . .
2.14. Climate indices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.14.1. ECACDD - Consecutive dry days index per time period . . . . . . . . . . .
2.14.2. ECACFD - Consecutive frost days index per time period . . . . . . . . . .
2.14.3. ECACSU - Consecutive summer days index per time period . . . . . . . . .
2.14.4. ECACWD - Consecutive wet days index per time period . . . . . . . . . . .
2.14.5. ECACWDI - Cold wave duration index wrt mean of reference period . . . .
2.14.6. ECACWFI - Cold-spell days index wrt 10th percentile of reference period .
2.14.7. ECAETR - Intra-period extreme temperature range . . . . . . . . . . . . .
2.14.8. ECAFD - Frost days index per time period . . . . . . . . . . . . . . . . . .
2.14.9. ECAGSL - Growing season length index . . . . . . . . . . . . . . . . . . . .
2.14.10.ECAHD - Heating degree days per time period . . . . . . . . . . . . . . . .
2.14.11.ECAHWDI - Heat wave duration index wrt mean of reference period . . . .
2.14.12.ECAHWFI - Warm spell days index wrt 90th percentile of reference period
2.14.13.ECAID - Ice days index per time period . . . . . . . . . . . . . . . . . . . .
2.14.14.ECAR10MM - Heavy precipitation days index per time period . . . . . . .
2.14.15.ECAR20MM - Very heavy precipitation days index per time period . . . .
2.14.16.ECAR75P - Moderate wet days wrt 75th percentile of reference period . . .
2.14.17.ECAR75PTOT - Precipitation percent due to R75p days . . . . . . . . . .
2.14.18.ECAR90P - Wet days wrt 90th percentile of reference period . . . . . . . .
2.14.19.ECAR90PTOT - Precipitation percent due to R90p days . . . . . . . . . .
2.14.20.ECAR95P - Very wet days wrt 95th percentile of reference period . . . . .
2.14.21.ECAR95PTOT - Precipitation percent due to R95p days . . . . . . . . . .

Contents
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

105
106
107
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
126
127
127
128
128
129
130
130
130
131
131
131
132
133
134
135
137
137
138
138
139
139
140
140
141
142
142
143
143
144
144
145
145
146
146
147
147

Contents
2.14.22.ECAR99P - Extremely wet days wrt 99th percentile of reference period . . .
2.14.23.ECAR99PTOT - Precipitation percent due to R99p days . . . . . . . . . . .
2.14.24.ECARR1 - Wet days index per time period . . . . . . . . . . . . . . . . . . .
2.14.25.ECARX1DAY - Highest one day precipitation amount per time period . . . .
2.14.26.ECARX5DAY - Highest five-day precipitation amount per time period . . . .
2.14.27.ECASDII - Simple daily intensity index per time period . . . . . . . . . . . .
2.14.28.ECASU - Summer days index per time period . . . . . . . . . . . . . . . . . .
2.14.29.ECATG10P - Cold days percent wrt 10th percentile of reference period . . .
2.14.30.ECATG90P - Warm days percent wrt 90th percentile of reference period . .
2.14.31.ECATN10P - Cold nights percent wrt 10th percentile of reference period . .
2.14.32.ECATN90P - Warm nights percent wrt 90th percentile of reference period . .
2.14.33.ECATR - Tropical nights index per time period . . . . . . . . . . . . . . . . .
2.14.34.ECATX10P - Very cold days percent wrt 10th percentile of reference period .
2.14.35.ECATX90P - Very warm days percent wrt 90th percentile of reference period
A. Hints for PINGO user

Contents
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.

148
148
149
149
150
150
150
152
152
153
153
154
154
155
157

B. Grid description examples

158
B.1. Example of a curvilinear grid description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
B.2. Example description for unstructured grid cells . . . . . . . . . . . . . . . . . . . . . . . . . 159
Operator index

160

1. Introduction
The Climate Data Operators (CDO) software are a collection of many operators for standard processing
of climate and forecast model output. The operators include simple statistical and arithmetic functions,
data selection and subsampling tools, and spatial interpolation. CDO was developed to have the same set
of processing functions for GRIB and netCDF datasets in one package.
The Climate Data Interface (CDI) is used for the fast and file format independent access to GRIB and
netCDF datasets. The local data formats SERVICE, EXTRA and IEG are also supported.
There are some limitations for GRIB and netCDF datasets. A GRIB dataset has to be consistent, similar
to netCDF. That means all time steps need to have the same variables, and within a time step each
variable may occur only once. NetCDF datasets are supported only with 2-dimensional, 3-dimensional and
4-dimensional variables and the attributes should follow the GDT, COARDS or CF Conventions.
The user interface and some operators are similar to the PINGO package. There are also some operators
with the same name as in PINGO but with a different meaning. Appendix A gives an overview of those
operators.
The main CDO features are:
More than 400 operators available
Modular design and easily extendable with new operators
Very simple UNIX command line interface
A dataset can be processed by several operators, without storing the interim results in files
Most operators handle datasets with missing values
Fast processing of large datasets
Support of many different grid types
Tested on many UNIX/Linux systems, Cygwin, and MacOS-X

1.1. Building from sources

This section describes how to build CDO from the sources on a UNIX system. CDO uses the GNU configure
and build system for compilation. The only requirement is a working ANSI C99 compiler.
First go to the download page (http://www.mpimet.mpg.de/cdo) to get the latest distribution, if you do
not have it yet.
To take full advantage of CDO features the following additional libraries should be installed:
Unidata netCDF library (http://www.unidata.ucar.edu/packages/netcdf) version 3 or higher.
This is needed to process netCDF files with CDO.
HDF5 szip library (http://www.hdfgroup.org/doc resource/SZIP) version 2.1 or higher.
This is needed to process szip compressed GRIB files with CDO.
HDF5 library (http://www.hdfgroup.org/HDF5) version 1.6 or higher.
This is needed to import CM-SAF HDF5 files with the CDO operator import cmsaf.
PROJ.4 library (http://trac.osgeo.org/proj) version 4.6 or higher.
This is needed to convert Sinusoidal and Lambert Azimuthal Equal Area coordinates to geographic
coordinates, for e.g. remapping.

Introduction

Building from sources

1.1.1. Compilation
Compilation is done by performing the following steps:
1. Unpack the archive, if you havent done that yet:
gunzip cdo-$VERSION.tar.gz
tar xf cdo-$VERSION.tar
cd cdo-$VERSION

# uncompress the archive

# unpack it

2. Run the configure script:

./configure
Or with netCDF support:
./configure --with-netcdf=<netCDF root directory>
The netCDF-4 configuration depends on the netCDF-4 and HDF5 installation! You have to define
the location of the HDF5 installation if netCDF-4 was build with HDF5 support:
./configure --with-netcdf=<netCDF-4 root directory> \
--with-hdf5=<HDF5 root directory>
You have to specify also the location of the SZLIB if HDF5 was build with SZLIB support.
To enable szip support add:
--with-szlib=<SZLIB root directory>
For an overview of other configuration options use
./configure --help
3. Compile the program by running make:
make
The program should compile without problems and the binary (cdo) should be available in the src directory
of the distribution.

1.1.2. Installation
After the compilation of the source code do a make install, possibly as root if the destination permissions
require that.
make install
The binary is installed into the directory <prefix>/bin. <prefix> defaults to /usr/local but can be
changed with the --prefix option of the configure script.
Alternatively, you can also copy the binary from the src directory manually to some bin directory in your
search path.

Usage

Introduction

1.2. Usage
This section descibes how to use CDO. The syntax is:
cdo

[ Options ] Operator1 [ -Operator2 [ -OperatorN ] ]

1.2.1. Options
All options have to be placed before the first operator. The following options are available for all operators:
-a
-b <nbits>

Convert from a relative to an absolute time axis.

Set the number of bits for the output precision. The valid precisions depend
on the file format:
<format>
grb
nc, nc2, nc4, srv, ext, ieg

-f <format>

For srv, ext and ieg format the letter L or B can be added to set the byteorder
to Little or Big endian.
Set the output file format. The valid file formats are:
File format
GRIB version 1
netCDF
netCDF version 2 (64-bit)
netCDF-4 classic (HDF5)
SERVICE
EXTRA
IEG

-g <grid>
-h
-m <missval>
-Q
-R
-r
-s
-t <partab>
-V
-v
-z szip
zip

<nbits>
1 - 32
32/64

<format>
grb
nc
nc2
nc4
srv
ext
ieg

Define the default grid description by name or from file (see chapter 1.3 on page 9).
Available grid names are: t<RES>grid, r<NX>x<NY>, gme<NI>
Help information for the operators.
Set the default missing value (default: -9e+33).
Sort netCDF variable names.
Convert GRIB data from reduced to regular grid.
Convert from an absolute to a relative time axis.
Silent mode.
Set the default parameter table name or file (see chapter 1.6 on page 14).
Predefined tables are: echam4 echam5 mpiom1
Print the version number.
Print extra details for some operators.
Compress GRIB records with szip.
Deflate compression of netCDF4 variables.

1.2.2. Operators
There are more than 400 operators available. A detailed description of all operators can be found in the
Reference Manual section.

1.2.3. Combining operators

All operators with a fixed number of input streams and one output stream can pipe the result directly
to an other operator. The operator must begin with , in order to combine it with others. This can
improve the performance by:

Introduction

Grid description

reducing unnecessary disk I/O

parallel processing

Use
cdo sub -dayavg ifile2 -timavg ifile1 ofile
instead of
cdo timavg ifile1 tmp1
cdo dayavg ifile2 tmp2
cdo sub tmp2 tmp1 ofile
rm tmp1 tmp2
Combining of operators is implemented over POSIX Threads (pthread). Therefore this CDO feature is not
available on operating systems without POSIX Threads support.

1.2.4. Operator parameter

Some operators need one or more parameter.
STRING

Unquoted characters without blanks and tabs. The following command select the variables with the
names pressure and tsurf:
cdo selvar,pressure,tsurf ifile ofile
FLOAT

Floating point number in any representation. The following command sets the range between 0 and
273.15 of all fields to missing value:
cdo setrtomiss,0,273.15 ifile ofile
INTEGER

A list of integers can be specified by first/last[/inc]. To select the days 5, 6, 7, 8 and 9 use:
cdo selday,5/9 ifile ofile
This is the same as:
cdo selday,5,6,7,8,9 ifile ofile

1.3. Grid description

In the following situations it is necessary to give a description of a horizontal grid:
Changing the grid description (operator: setgrid)
Horizontal interpolation (operator: interpolate, remapXXX and genXXX)
Generating variables (operator: const, random)

As now described, there are several possibilities to define a horizontal grid. Predefined grids are available
for global regular, gaussian or icosahedral-hexagonal GME grids.

Grid description

Introduction

1.3.1. Predefined grids

The following pre-defined grid names are available: r<NX>x<NY>, lon=<LON> lat<LAT>, t<RES>grid and
gme<NI>

Global regular grid: r<NX>x<NY>

r<NX>x<NY> defines a global regular lon/lat grid. The number of the longitudes <NX> and the latitudes
<NY> can be selected at will. The longitudes start at 0 with an increment of (360/<NX>) . The latitudes
go from south to north with an increment of (180/<NY>) .

One grid point: lon=<LON> lat=<LAT>

lon=<LON> lat=<LAT> defines one grid point of a lon/lat grid.

Global gaussian grid: t<RES>grid

t<RES>grid defines a global gaussian grid. Each valid triangular resolution can be used for <RES>. The
longitudes start at 0 with an increment of (360/nlon) . The gaussian latitudes go from north to south.

Global icosahedral-hexagonal GME grid: gme<NI>

gme<NI> defines a global icosahedral-hexagonal GME grid. NI is the number of intervals on a main triangle
side.

1.3.2. Grids from data files

You can use the grid description from an other datafile. The format of the datafile and the grid of the
data field must be supported by this program. Use the operator sinfo to get short informations about
your variables and the grids. If there are more then one grid in the datafile the grid description of the first
variable will be used.

1.3.3. SCRIP grids

SCRIP is a Spherical Coordinate Remapping and Interpolation Package. It uses a common grid description
in netCDF. You can use it to describe curvilinear grids or unstructured grid cells. For more information
about this format see [SCRIP]. That grid description format is only available if the program was compiled
with netCDF support.
SCRIP grid description example of a curvilinear MPIOM GROB3 grid (only the netCDF header):
netcdf grob3s {
dimensions :
g r i d s i z e = 12120 ;
g r i d x s i z e = 120 ;
g r i d y s i z e = 101 ;
grid corners = 4 ;
grid rank = 2 ;
variables :
int grid dims ( grid rank ) ;
float grid center lat ( grid ysize , grid xsize )
g r i d c e n t e r l a t : units = degrees ;
g r i d c e n t e r l a t : bounds = g r i d c o r n e r
float grid center lon ( grid ysize , grid xsize )
grid center lon : units = degrees ;
g r i d c e n t e r l o n : bounds = g r i d c o r n e r

10

;
lat ;
;
lon ;

Introduction

Grid description
int grid imask ( grid ysize , g r i d x s i z e ) ;
grid imask : units = u n i t l e s s ;
grid imask : coordinates = grid center lon g r i d c e n t e r l a t ;
float grid corner lat ( grid ysize , grid xsize , grid corners ) ;
g r i d c o r n e r l a t : units = degrees ;
float grid corner lon ( grid ysize , grid xsize , grid corners ) ;
grid corner lon : units = degrees ;

// g l o b a l a t t r i b u t e s :
: t i t l e = grob3s ;
}

1.3.4. PINGO grids

PINGO uses a very simple grid description in ASCII format to describe regular longitude/latitude or global
gaussian grids. All PINGO grid description files are supported by CDO. For more information about this
format see [PINGO].
PINGO grid description example of a T21 gaussian grid:
Grid D e s c r i p t i o n F i l e
( Comments s t a r t a t non d i g i t c h a r a c t e r s and end a t end o f l i n e )
F i r s t p a r t : The d i m e n s i o n s .
64 32 = Number o f l o n g i t u d e s and l a t i t u d e s
Second p a r t : The l i s t e d l o n g i t u d e s .
2 means e q u i d i s t a n t l o n g i t u d e s
0 . 0 0 0 0 0 0 5 . 6 2 5 0 0 0 = Most w e s t e r n and s e c o n d most w e s t e r n l o n g i t u d e
Third p a r t : The l i s t e d l a t i t u d e s .
32 means a l l 32 l a t i t u d e s a r e g i v e n i n t h e f o l l o w i n g l i s t :
85.761
80.269
74.745
69.213
63.679
58.143
52.607
47.070
41.532
35.995
30.458
24.920
19.382
13.844
8.307
2.769
2.769 8.307 13.844 19.382 24.920 30.458 35.995 41.532
47.070 52.607 58.143 63.679 69.213 74.745 80.269 85.761

1.3.5. CDO grids

All supported grids can also be described with the CDO grid description. The following keywords can be
used to describe a grid:
Keyword
gridtype
gridsize
xsize
ysize
xvals
yvals
xnpole
ynpole
nvertex
xbounds
ybounds
xfirst, xinc
yfirst, yinc

Datatype
STRING
INTEGER
INTEGER
INTEGER
FLOAT ARRAY
FLOAT ARRAY
FLOAT
FLOAT
INTEGER
FLOAT ARRAY
FLOAT ARRAY
FLOAT, FLOAT
FLOAT, FLOAT

Description
type of the grid (gaussian, lonlat, curvilinear, cell)
size of the grid
size in x direction (number of longitudes)
size in y direction (number of latitudes)
x values of the grid
y values of the grid
x value of the north pole (rotated grid)
y value of the north pole (rotated grid)
number of the vertices for all grid cells
x bounds of each gridbox
y bounds of each gridbox
macros to define xvals with a constant increment
macros to define yvals with a constant increment

Which keywords are necessary depends on the gridtype. The following table gives an overview of the
default values or the size with respect to the different grid types.

11

Z-axis description

Introduction

gridtype

lonlat

gaussian

curvilinear

cell

gridsize
xsize
ysize
xvals
yvals
xnpole
ynpole
nvertex
xbounds
ybounds

xsize*ysize
nlon
nlat
xsize
ysize
0
90
2
2*xsize
2*ysize

xsize*ysize
nlon
nlat
xsize
ysize

xsize*ysize
nlon
nlat
gridsize
gridsize

ncell
gridsize
gridsize
gridsize
gridsize

2
2*xsize
2*ysize

4
4*gridsize
4*gridsize

nv
nv*gridsize
nv*gridsize

The keywords nvertex, xbounds and ybounds are optional if area weights are not needed.
CDO grid description example of a T21 gaussian grid:
gridtype
xsize
ysize
xfirst
xinc
yvals

=
=
=
=
=
=

gaussian
64
32
0
5.625
85.76
80.27
74.75
69.21
63.68
58.14
52.61
47.07
41.53
36.00
30.46
24.92
19.38
13.84
8.31
2.77
2.77 8.31 13.84 19.38 24.92 30.46 36.00 41.53
47.07 52.61 58.14 63.68 69.21 74.75 80.27 85.76

CDO grid description example of a global regular grid with 60x30 points:
gridtype
xsize
ysize
xfirst
xinc
yfirst
yinc

= lonlat
=
60
=
30
= 177
=
6
= 87
=
6

For a lon/lat grid with a rotated pole, the north pole must be defined. As far as you define the keywords
xnpole/ynpole all coordinate values are for the rotated system.
CDO grid description example of a regional rotated lon/lat grid:
gridtype
xsize
ysize
xfirst
xinc
yfirst
yinc
xnpole
ynpole

=
=
=
=
=
=
=
=
=

lonlat
81
91
19.5
0.5
25.0
0.5
170
32.5

Example CDO descriptions of a curvilinear and an unstructured grid can be found in Appendix B.

1.4. Z-axis description

Sometimes it is necessary to change the description of a z-axis. This can be done with the operator setzaxis.
This operator needs an ASCII formatted file with the description of the z-axis. The following keywords
can be used to describe a z-axis:

12

Introduction
Keyword
zaxistype
size
levels
lbounds
ubounds
vctsize
vct

Time axis
Datatype
STRING
INTEGER
FLOAT ARRAY
FLOAT ARRAY
FLOAT ARRAY
INTEGER
FLOAT ARRAY

Description
type of the z-axis
number of levels
values of the levels
lower level bounds
upper level bounds
number of vertical coordinate parameters
vertical coordinate table

The keywords lbounds and ubounds are optional. vctsize and vct are only necessary to define hybrid
model levels.
Available z-axis types:
Z-axis type
surface
pressure
hybrid
height
depth below sea
depth below land
isentropic

Description
Surface
Pressure level
Hybrid model level
Height above ground
Depth below sea level
Depth below land surface
Isentropic (theta) level

Units
pascal
meter
meter
centimeter
kelvin

Z-axis description example for pressure levels 100, 200, 500, 850 and 1000 hPa:
zaxistype = pressure
size
= 5
levels
= 10000 20000 50000 85000 100000

Z-axis description example for ECHAM5 L19 hybrid model levels:

zaxistype
size
levels
vctsize
vct

=
=
=
=
=

hybrid
19
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
40
0 2000 4000 6 0 4 6 . 1 0 9 3 8 8 2 6 7 . 9 2 5 7 8 1 0 6 0 9 . 5 1 1 7 1 2 8 5 1 . 1 0 1 6 1 4 6 9 8 . 5
15861.125 16116.2383 15356.9258 13621.4609 11101.5625 8127.14453
5125.14062 2549.96875 783.195068 0 0 0
0 0 0 0.000338993268 0.00335718691 0.0130700432 0.0340771675
0.0706498027 0.12591666 0.201195419 0.295519829 0.405408859
0.524931908 0.646107674 0.759697914 0.856437683 0.928747177
0.972985268 0.992281914 1

1.5. Time axis

A time axis describes the time for every timestep. Two time axis types are available: absolute time and
relative time axis. CDO tries to maintain the actual type of the time axis for all operators. Some time range
statistic operators create an absolute time axis (e.g. timsel< ST AT >, run< ST AT >, seas< ST AT >, ...).

1.5.1. Absolute time

An absolute time axis has the current time to each time step. It can be used without knowledge of the
calendar. This is preferably used by climate models. In netCDF files the absolute time axis is represented
by the unit of the time: day as %Y%m%d.%f.

13

Parameter table

Introduction

1.5.2. Relative time

A relative time is the time relative to a fixed reference time. The current time results from the reference
time and the elapsed interval. The result depends on the calendar used. CDO supports the standard
Gregorian, 360 days, 365 days and 366 days calendars. The relative time axis is preferably used by weather
forecast models. In netCDF files the relative time axis is represented by the unit of the time: time-units
since reference-time, e.g days since 1989-6-15 12:00.

1.5.3. Conversion of the time

Some programs which work with netCDF data can only process relative time axes. Therefore it may be
necessary to convert from an absolute into a relative time axis. This conversion can be done for each
operator with the CDO option -r. To convert a relative into an absolute time axis use the CDO option
-a.

1.6. Parameter table

A parameter table is an ASCII formated file to convert code numbers to variable names. Each variable
has one line with its code number, name and a description with optional units in a blank separated list. It
can only be used for GRIB, SERVICE, EXTRA and IEG formated files. The CDO option -t <partab>
sets the default parameter table for all input files. Use the operator setpartab to set the parameter table
for a specific file.
Example of a CDO parameter table:
134
141
147
172
175
211

aps
sn
ahfl
slm
albedo
siced

s u r f a c e p r e s s u r e [ Pa ]
snow depth [m]
l a t e n t h e a t f l u x [W/m* * 2 ]
l a n d s e a mask
s u r f a c e albedo
i c e depth [m]

1.7. Missing values

Most operators can handle missing values. The default missing value for GRIB, SERVICE, EXTRA and
IEG files is 9e + 33. The CDO option -m <missval> overwrites the default missing value. In netCDF
files the variable attribute FillValue is used as a missing value. The operator setmissval can be used to
set a new missing value.
The CDO use of the missing value is shown in the following tables, where one table is printed for each
operation. The operations are applied to arbitrary numbers a, b, the special case 0, and the missing
value miss. For example the table named addition shows that the sum of an arbitrary number a and
the missing value is the missing value, and the table named multiplication shows that 0 multiplied by
missing value results in 0.

14

Introduction

Missing values

addition

miss

a
miss

a+b
miss

miss
miss

subtraction

miss

a
miss

ab
miss

miss
miss

multiplication

miss

a
0
miss

ab
0
miss

0
0
0

miss
0
miss

division

miss

a
0
miss

a/b
0
miss

miss
miss
miss

miss
miss
miss

maximum

miss

a
miss

max(a, b)
b

a
miss

minimum

miss

a
miss

min(a, b)
b

a
miss

sum

miss

a
miss

a+b
b

a
miss

The handling of missing values by the operations minimum and maximum may be surprising, but the
definition given here is more consistent with that expected in practice. Mathematical functions (e.g. log,
sqrt, etc.) return the missing value if an argument is the missing value or an argument is out of range.
All statistical functions ignore missing values, treading them as not belonging to the sample, with the
side-effect of a reduced sample size.

1.7.1. Mean and average

An artificial distinction is made between the notions mean and average. The mean is regarded as a
statistical function, whereas the average is found simply by adding the sample members and dividing the
result by the sample size. For example, the mean of 1, 2, miss and 3 is (1 + 2 + 3)/3 = 2, whereas the
average is (1 + 2 + miss + 3)/4 = miss/4 = miss. If there are no missing values in the sample, the average
and mean are identical.

15

2. Reference manual
This section gives a description of all operators. Related operators are grouped to modules. For easier
description all single input files are named ifile or ifile1, ifile2, etc., and an unlimited number of
input files are named ifiles. All output files are named ofile or ofile1, ofile2, etc. Further the
following notion is introduced:
i(t)

Timestep t of ifile

i(t, x)

Element number x of the field at timestep t of ifile

o(t)

Timestep t of ofile

o(t, x)

Element number x of the field at timestep t of ofile

16

Reference manual

Information

2.1. Information
This section contains modules to print information about datasets. All operators print there results to
standard output.
Here is a short overview of all operators in this section:
info
infov
map

Dataset information listed by code number

Dataset information listed by variable name
Dataset information and simple map

sinfo
sinfov

Short dataset information listed by code number

Short dataset information listed by variable name

diff
diffv

Compare two datasets listed by code number

Compare two datasets listed by variable name

npar
nlevel
nyear
nmon
ndate
ntime

Number
Number
Number
Number
Number
Number

showformat
showcode
showname
showstdname
showlevel
showltype
showyear
showmon
showdate
showtime

Show
Show
Show
Show
Show
Show
Show
Show
Show
Show

pardes
griddes
zaxisdes
vct

Parameter description
Grid description
Z-axis description
Vertical coordinate table

of
of
of
of
of
of

parameters
levels
years
months
dates
time steps

file format
code numbers
variable names
standard names
levels
GRIB level types
years
months
dates
time steps

17

Information

Reference manual

2.1.1. INFO - Information and simple statistics

Synopsis
< operator > ifiles

Description
This module writes information about the structure and contents of all input datasets to standard
output. The information displayed depends on the chosen operator.

Operators
info

Dataset information listed by code number

Prints information and simple statistics for each field of all input datasets. For each field
the operator prints one line with the following elements:
Date and Time
Code number and Level
Size of the grid and number of Missing values
Minimum, Mean and Maximum
The mean value is computed without the use of area weights!

infov

Dataset information listed by variable name

The same as operator info but using the name instead of the code number to identify the
variables.

map

Dataset information and simple map

Prints information, simple statistics and a map for each field of all input datasets. The
map will be printed only for fields on a regular lon/lat grid.

Example
To print information and simple statistics for each field of a dataset use:
cdo i n f o

ifile

This is an example result of a dataset with one 2D variable over 12 time steps:
1
1
2
3
4
5
6
7
8
9
10
11
12

18

:
:
:
:
:
:
:
:
:
:
:
:
:

Date
19870131
19870228
19870331
19870430
19870531
19870630
19870731
19870831
19870930
19871031
19871130
19871231

Time
12:00:00
12:00:00
12:00:00
12:00:00
12:00:00
12:00:00
12:00:00
12:00:00
12:00:00
12:00:00
12:00:00
12:00:00

Code L e v e l
139
0
139
0
139
0
139
0
139
0
139
0
139
0
139
0
139
0
139
0
139
0
139
0

Size
2048
2048
2048
2048
2048
2048
2048
2048
2048
2048
2048
2048

Miss
1361
1361
1361
1361
1361
1361
1361
1361
1361
1361
1361
1361

:
:
:
:
:
:
:
:
:
:
:
:
:

Minimum
232.77
233.64
225.31
215.68
215.78
212.89
209.52
210.48
210.48
219.46
230.98
241.25

Mean
266.65
267.11
267.52
268.65
271.53
272.80
274.29
274.41
272.37
270.53
269.85
269.94

Maximum
305.31
307.15
307.67
310.47
312.49
314.18
316.34
315.83
312.86
309.51
308.61
309.27

Reference manual

Information

2.1.2. SINFO - Short information

Synopsis
< operator > ifiles

Description
This module writes information about the structure of all input datasets to standard output. The
information displayed depends on the chosen operator.

Operators
sinfo

Short dataset information listed by code number

Prints short information of a dataset. The information is divided into 4 sections. Section
1 prints one line per variable with the following information:
institute and source
parameter table and code number
horizontal grid size and number
number of vertical levels and z-axis number

Section 2 and 3 gives a short overview of all horizontal and vertical grids. And the last
section contains short information of the time axis.
sinfov

Short dataset information listed by variable name

The same as operator sinfo but using the name instead of the code number and parameter
table to identify the variables.

Example
To print short information of a dataset use:
cdo s i n f o

ifile

This is the result of an ECHAM5 dataset with 3 variables over 12 time steps:
1 : I n s t i t u t S o u r c e Table Code
Time
Typ Grid S i z e Num L e v e l s Num
1 : MPIMET
ECHAM5. 3 128 129 c o n s t a n t F32
2048
1
1
1
2 : MPIMET
ECHAM5. 3 128 130 v a r i a b l e F32
2048
1
4
2
3 : MPIMET
ECHAM5. 3 128 139 v a r i a b l e F32
2048
1
1
1
Horizontal grids :
1 : gaussian
> size
: dim = 2048 n l o n = 64 n l a t = 32
longitude : f i r s t = 0 l a s t = 354.375 inc = 5.625
latitude
: f i r s t = 85.7605871
l a s t = 85.7605871
Vertical grids :
1 : surface
: 0
2 : pressure
Pa : 92500 85000 50000 20000
Time a x i s : 12 s t e p s
YYYYMMDD hh :mm: s s YYYYMMDD hh :mm: s s YYYYMMDD hh :mm: s s YYYYMMDD hh :mm: s s
19870131 1 2 : 0 0 : 0 0 19870228 1 2 : 0 0 : 0 0 19870331 1 2 : 0 0 : 0 0 19870430 1 2 : 0 0 : 0 0
19870531 1 2 : 0 0 : 0 0 19870630 1 2 : 0 0 : 0 0 19870731 1 2 : 0 0 : 0 0 19870831 1 2 : 0 0 : 0 0
19870930 1 2 : 0 0 : 0 0 19871031 1 2 : 0 0 : 0 0 19871130 1 2 : 0 0 : 0 0 19871231 1 2 : 0 0 : 0 0

19

Information

Reference manual

2.1.3. DIFF - Compare two datasets field by field

Synopsis
< operator > ifile1 ifile2

Description
Compares the contents of two datasets field by field. The input datasets need to have the same
structure and its fields need to have the same header information and dimensions.

Operators
diff

Compare two datasets listed by code number

Provides statistics on differences between two datasets. For each pair of fields the operator
prints one line with the following information:
Date and Time
Code number and Level
Size of the grid and number of Missing values
Occurrence of coefficient pairs with different signs (S)
Occurrence of zero values (Z)
Maxima of absolute difference of coefficient pairs
Maxima of relative difference of non-zero coefficient pairs with equal signs

Absdif f (t, x) = |i1 (t, x) i2 (t, x)|

Reldif f (t, x) =
diffv

|i1 (t, x) i2 (t, x)|

max(|i1 (t, x)| , |i2 (t, x)|)

Compare two datasets listed by variable name

The same as operator diff. Using the name instead of the code number to identify the
variable.

Example
To print the difference for each field of two datasets use:
cdo d i f f

ifile1

ifile2

This is an example result of two datasets with one 2D variable over 12 time steps:
1
2
3
4
5
6
7
8
9
10
11
12

20

:
:
:
:
:
:
:
:
:
:
:
:

Date
19870131
19870228
19870331
19870430
19870531
19870630
19870731
19870831
19870930
19871031
19871130
19871231

Time
12:00:00
12:00:00
12:00:00
12:00:00
12:00:00
12:00:00
12:00:00
12:00:00
12:00:00
12:00:00
12:00:00
12:00:00

Code L e v e l
139
0
139
0
139
0
139
0
139
0
139
0
139
0
139
0
139
0
139
0
139
0
139
0

Size
2048
2048
2048
2048
2048
2048
2048
2048
2048
2048
2048
2048

Miss
1361
1361
1361
1361
1361
1361
1361
1361
1361
1361
1361
1361

:
:
:
:
:
:
:
:
:
:
:
:
:

S
F
F
F
F
F
F
F
F
F
F
F
F

Z Max Absdiff M a x R e l d i f f
F 0 . 0 0 0 1 0 6 8 1 4 . 1 6 6 0 e 07
F 6 . 1 0 3 5 e 05 2 . 3 7 4 2 e 07
F 7 . 6 2 9 4 e 05 3 . 3 7 8 4 e 07
F 7 . 6 2 9 4 e 05 3 . 5 1 1 7 e 07
F 0 . 0 0 0 1 0 6 8 1 4 . 0 3 0 7 e 07
F 0 . 0 0 0 1 0 6 8 1 4 . 2 6 7 0 e 07
F 9 . 1 5 5 3 e 05 3 . 5 6 3 4 e 07
F 7 . 6 2 9 4 e 05 2 . 8 8 4 9 e 07
F 7 . 6 2 9 4 e 05 3 . 6 1 6 8 e 07
F 9 . 1 5 5 3 e 05 3 . 5 0 0 1 e 07
F 6 . 1 0 3 5 e 05 2 . 3 8 3 9 e 07
F 9 . 3 5 5 3 e 05 3 . 7 6 2 4 e 07

Reference manual

Information

2.1.4. NINFO - Print the number of parameters, levels or times

Synopsis
< operator > ifile

Description
This module prints the number of variables, levels or times of the input dataset.

Operators
npar

Number of parameters
Prints the number of parameters (variables).

nlevel

Number of levels
Prints the number of levels for each variable.

nyear

Number of years
Prints the number of different years.

nmon

Number of months
Prints the number of different combinations of years and months.

ndate

Number of dates
Prints the number of different dates.

ntime

Number of time steps

Prints the number of time steps.

Example
To print the number of parameters (variables) in a dataset use:
cdo npar i f i l e

To print the number of months in a dataset use:

cdo nmon i f i l e

21

Information

Reference manual

2.1.5. SHOWINFO - Show variables, levels or times

Synopsis
< operator > ifile

Description
This module prints the format, variables, levels or times of the input dataset.

Operators
showformat

Show file format

Prints the file format of the input dataset.

showcode

Show code numbers

Prints the code number of all variables.

showname

Show variable names

Prints the name of all variables.

showstdname

Show standard names

Prints the standard name of all variables.

showlevel

Show levels
Prints all levels for each variable.

showltype

Show GRIB level types

Prints the GRIB level type for all z-axes.

showyear

Show years
Prints all years.

showmon

Show months
Prints all months.

showdate

Show dates
Prints all dates.

showtime

Show time steps

Prints all time steps.

Example
To print the code number of all variables in a dataset use:
cdo showcode i f i l e

This is an example result of a dataset with three variables:

129 130 139

To print all months in a dataset use:

cdo showmon i f i l e

This is an examples result of a dataset with an annual cycle:

1 2 3 4 5 6 7 8 9 10 11 12

22

Reference manual

Information

2.1.6. FILEDES - Dataset description

Synopsis
< operator > ifile

Description
This module prints the description of the parameters, the grids, the z-axis or the vertical coordinate
table.

Operators
pardes

Parameter description
Prints a table with a description of all variables. For each variable the operator prints
one line listing the code, name, description and units.

griddes

Grid description
Prints the description of all grids.

zaxisdes

Z-axis description
Prints the description of all z-axes.

vct

Vertical coordinate table

Prints the vertical coordinate table.

Example
Assume all variables of the dataset are on a T21 gausssian grid. To print the grid description of this
dataset use:
cdo g r i d d e s i f i l e

Result:
gridtype
gridsize
xname
xlongname
xunits
yname
ylongname
yunits
xsize
ysize
xfirst
xinc
yvals

:
:
:
:
:
:
:
:
:
:
:
:
:

gaussian
2048
lon
longitude
degrees east
lat
latitude
degrees north
64
32
0
5.625
85.76058 80.26877 74.74454 69.21297 63.67863 58.1429 52.6065
47.06964 41.53246 35.99507 30.4575 24.91992 19.38223 13.84448
8 . 3 0 6 7 0 2 2 . 7 6 8 9 0 3 2.768903 8.306702 13.84448 19.38223
24.91992 30.4575 35.99507 41.53246 47.06964 52.6065
58.1429 63.67863 69.21297 74.74454 80.26877 85.76058

23

File operations

Reference manual

2.2. File operations

This section contains modules to perform operations on files.
Here is a short overview of all operators in this section:

24

copy
cat

Copy datasets
Concatenate datasets

replace

Replace variables

merge
mergetime

Merge datasets with different fields

Merge datasets sorted by date and time

splitcode
splitname
splitlevel
splitgrid
splitzaxis

Split
Split
Split
Split
Split

code numbers
variable names
levels
grids
z-axes

splithour
splitday
splitmon
splitseas
splityear

Split
Split
Split
Split
Split

hours
days
months
seasons
years

splitsel

Split time selection

Reference manual

File operations

2.2.1. COPY - Copy datasets

Synopsis
< operator > ifiles ofile

Description
This module contains operators to copy or concatenate datasets. Each input dataset is required to
have the same variables with complete time steps.

Operators
copy

Copy datasets
Copies all input datasets to ofile.

cat

Concatenate datasets
Concatenates all input datasets and appends the result to the end of ofile. If ofile does
not exist it will be created.

Example
To change the format of a dataset to netCDF use:
cdo f nc copy i f i l e

o f i l e . nc

Add the option -r to create a relative time axis, as is required for proper recognition by GrADS or
Ferret:
cdo r f nc copy i f i l e

o f i l e . nc

To concatenate 3 datasets with different time steps of the same variables use:
cdo copy i f i l e 1

ifile2

ifile3

ofile

If the output dataset already exists and you wish to extend it with more time steps use:
cdo c a t i f i l e 1

ifile2

ifile3

ofile

2.2.2. REPLACE - Replace variables

Synopsis
replace ifile1 ifile2 ofile

Description
Replaces all common variables of ifile2 and ifile1 with those of ifile1 and write the result to
ofile. Both input datasets need to have the same number of time steps.

Example
Assume the first input dataset ifile1 has three variables with the names geosp, t and tslm1 and the
second input dataset ifile2 has only the variable tslm1. To replace the variable tslm1 in ifile1
with tslm1 from ifile2 use:
cdo r e p l a c e i f i l e 1

ifile2

ofile

25

File operations

Reference manual

2.2.3. MERGE - Merge datasets

Synopsis
< operator > ifiles ofile

Description
This module reads datasets from several input files, merges them and writes the resulting dataset to
ofile.

Operators
merge

Merge datasets with different fields

Merges time series of different fields from several input datasets. The number of
fields per time step written to ofile is the sum of the field numbers per time step in
all input datasets. The time series on all input datasets are required to have different
fields and the same number of time steps.

mergetime

Merge datasets sorted by date and time

Merges all time steps of all input files sorted by date and time. After this operation
every input time step is in ofile and all time steps are sorted by date and time.
Each input file is required to have the same variables and different time steps.

Example
Assume three datasets with the same number of time steps and different variables in each dataset.
To merge these datasets to a new dataset use:
cdo merge i f i l e 1

ifile2

ifile3

ofile

Assume you split a 6 hourly dataset with splithour. This produces four datasets, one for each hour.
The following command merges them together:
cdo mergetime i f i l e 1

26

ifile2

ifile3

ifile4

ofile

Reference manual

File operations

2.2.4. SPLIT - Split a dataset

Synopsis
< operator > ifile oprefix

Description
This module splits a dataset to several files with names formed from the field header information and
the string oprefix.

Operators
splitcode

Split code numbers

Splits a dataset into pieces, one for each different code number. Appends three digits
with the code number to oprefix to form the output file names.

splitname

Split variable names

Splits a dataset into pieces, one for each variable name. Appends a string with the
variable name to oprefix to form the output file names.

splitlevel

Split levels
Splits a dataset into pieces, one for each different level. Appends six digits with the
level to oprefix to form the output file names.

splitgrid

Split grids
Splits a dataset into pieces, one for each different grid. Appends two digits with the
grid number to oprefix to form the output file names.

splitzaxis

Split z-axes
Splits a dataset into pieces, one for each different z-axis. Appends two digits with the
z-axis number to oprefix to form the output file names.

Example
Assume an input GRIB dataset with three variables, e.g. code number 129, 130 and 139. To split
this dataset into three pieces, one for each code number use:
cdo s p l i t c o d e

i f i l e code

Result of dir code*:

code129 . grb code130 . grb code139 . grb

27

File operations

Reference manual

2.2.5. SPLITTIME - Split time steps of a dataset

Synopsis
< operator > ifile oprefix

Description
This module splits time steps of a dataset to several files with names formed from the field header
information and the string oprefix.

Operators
splithour

Split hours
Splits a file into pieces, one for each different hour. Appends two digits with the hour
to oprefix to form the output file names.

splitday

Split days
Splits a file into pieces, one for each different day. Appends two digits with the day to
oprefix to form the output file names.

splitmon

Split months
Splits a file into pieces, one for each different month. Appends two digits with the
month to oprefix to form the output file names.

splitseas

Split seasons
Splits a file into pieces, one for each different season. Appends three characters with
the season to oprefix to form the output file names.

splityear

Split years
Splits a file into pieces, one for each different year. Appends four digits with the year
to oprefix to form the output file names.

Example
Assume the input GRIB dataset has time steps from January to December. To split each month with
all variables into one separate file use:
cdo s p l i t m o n i f i l e mon

Result of dir mon*:

mon01 . grb
mon07 . grb

28

mon02 . grb
mon08 . grb

mon03 . grb
mon09 . grb

mon04 . grb
mon10 . grb

mon05 . grb
mon11 . grb

mon06 . grb
mon12 . grb

Reference manual

File operations

2.2.6. SPLITSEL - Split selected time steps

Synopsis
splitsel,nsets[,noffset[,nskip]] ifile oprefix

Description
This operator splits a dataset into pieces, one for each adjacent sequence t1 , ...., tn of time steps of
the same selected time range. Appends three digits with the sequence number to oprefix to form
the output file names.

Parameter
nsets

INTEGER

Number of input time steps for each output file

noffset

INTEGER

Number of input time steps skipped before the first time step range (optional)

nskip

INTEGER

Number of input time steps skipped between time step ranges (optional)

29

Selection

Reference manual

2.3. Selection
This section contains modules to select time steps, fields or a part of a field from a dataset.
Here is a short overview of all operators in this section:

30

selcode
delcode
selname
delname
selstdname
sellevel
sellevidx
selgrid
selgridname
selzaxis
selzaxisname
selltype
seltabnum

Select variables by code number

Delete variables by code number
Select variables by name
Delete variables by name
Select variables by standard name
Select levels
Select levels by index
Select grids
Select grids by name
Select z-axes
Select z-axes by name
Select GRIB level types
Select parameter table numbers

seltimestep
seltime
selhour
selday
selmon
selyear
selseas
seldate
selsmon

Select
Select
Select
Select
Select
Select
Select
Select
Select

sellonlatbox
selindexbox

Select a longitude/latitude box

Select an index box

time steps
times
hours
days
months
years
seasons
dates
single month

Reference manual

Selection

2.3.1. SELECT - Select fields

Synopsis

selcode,codes ifile ofile

delcode,codes ifile ofile

selname,varnames ifile ofile

delname,varnames ifile ofile

selstdname,stdnames ifile ofile

sellevel,levels ifile ofile

sellevidx,levidx ifile ofile

selgrid,grids ifile ofile

selgridname,gridnames ifile ofile

selzaxis,zaxes ifile ofile

selzaxisname,zaxisnames ifile ofile

selltype,ltypes ifile ofile

seltabnum,tabnums ifile ofile

Description

This module selects some fields from ifile and writes them to ofile. The fields selected depend on
the chosen operator and the parameters.

31

Selection

Reference manual

Operators
selcode

Select variables by code number

Selects all fields with code numbers in a user given list.

delcode

Delete variables by code number

Deletes all fields with code numbers in a user given list.

selname

Select variables by name

Selects all fields with variable names in a user given list.

delname

Delete variables by name

Deletes all fields with variable names in a user given list.

selstdname

Select variables by standard name

Selects all fields with standard names in a user given list.

sellevel

Select levels
Selects all fields with levels in a user given list.

sellevidx

Select levels by index

Selects all fields with index of levels in a user given list.

selgrid

Select grids
Selects all fields with grids in a user given list.

selgridname

Select grids by name

Selects all fields with grid names in a user given list.

selzaxis

Select z-axes
Selects all fields with z-axes in a user given list.

selzaxisname

Select z-axes by name

Selects all fields with z-axis names in a user given list.

selltype

Select GRIB level types

Selects all fields with GRIB level type in a user given list.

seltabnum

Select parameter table numbers

Selects all fields with parameter table numbers in a user given list.

Parameter

32

codes

INTEGER

Comma separated list of code numbers

varnames

STRING

Comma separated list of variable names

stdnames

STRING

Comma separated list of standard names

levels

FLOAT

Comma separated list of levels

levidx

INTEGER

Comma separated list of index of levels

ltypes

INTEGER

Comma separated list of GRIB level types

grids

INTEGER

Comma separated list of grid numbers

gridnames

STRING

Comma separated list of grid names

zaxes

INTEGER

Comma separated list of z-axis numbers

zaxisnames

STRING

Comma separated list of z-axis names

tabnums

INTEGER

Comma separated list of parameter table numbers

Reference manual

Selection

2.3.2. SELTIME - Select time steps

Synopsis
seltimestep,timesteps ifile ofile
seltime,times ifile ofile
selhour,hours ifile ofile
selday,days ifile ofile
selmon,months ifile ofile
selyear,years ifile ofile
selseas,seasons ifile ofile
seldate,date1[,date2] ifile ofile
selsmon,month[,nts1[,nts2]] ifile ofile

Description
This module selects user specified time steps from ifile and writes them to ofile. The time steps
selected depends on the chosen operator and the parameters.

Operators
seltimestep

Select time steps

Selects all time steps with a time step in a user given list.

seltime

Select times
Selects all time steps with a time in a user given list.

selhour

Select hours
Selects all time steps with a hour in a user given list.

selday

Select days
Selects all time steps with a day in a user given list.

selmon

Select months
Selects all time steps with a month in a user given list.

selyear

Select years
Selects all time steps with a year in a user given list.

selseas

Select seasons
Selects all time steps with a month of a season in a user given list.

seldate

Select dates
Selects all time steps with a date in a user given range.

selsmon

Select single month

Selects a month and optional an unlimited number of time steps before and after
this month.

33

Selection

Reference manual

Parameter

34

timesteps

INTEGER

Comma separated list of time steps

times

STRING

Comma separated list of times (format hh:mm:ss)

hours

INTEGER

Comma separated list of hours

days

INTEGER

Comma separated list of days

months

INTEGER

Comma separated list of months

years

INTEGER

Comma separated list of years

seasons

STRING

Comma separated list of seasons (DJF, MAM, JJA, SON)

date1

STRING

Start date (format YYYY-MM-DDThh:mm:ss)

date2

STRING

End date (format YYYY-MM-DDThh:mm:ss)

nts1

INTEGER

Number of time steps before the selected month [default: 0]

nts2

INTEGER

Number of time steps after the selected month [default: nts1]

Reference manual

Selection

2.3.3. SELBOX - Select a box of a field

Synopsis
sellonlatbox,lon1,lon2,lat1,lat2 ifile ofile
selindexbox,idx1,idx2,idy1,idy2 ifile ofile

Description
Selects a box of the rectangular understood field. All input fields need to have the same horizontal
grid.

Operators
sellonlatbox

Select a longitude/latitude box

Selects a longitude/latitude box. The user has to give the longitudes and latitudes
of the edges of the box.

selindexbox

Select an index box

Selects an index box. The user has to give the indexes of the edges of the box. The
index of the left edge may be greater then that of the right edge.

Parameter
lon1

FLOAT

Western longitude

lon2

FLOAT

Eastern longitude

lat1

FLOAT

Southern or northern latitude

lat2

FLOAT

Northern or southern latitude

idx1

INTEGER

Index of first longitude

idx2

INTEGER

Index of last longitude

idy1

INTEGER

Index of first latitude

idy2

INTEGER

Index of last latitude

Example
To select the region with the longitudes from 120E to 90W and latitudes from 20N to 20S from all
input fields use:
cdo s e l l o n l a t b o x ,120 , 90 ,20 , 20 i f i l e

ofile

If the input dataset has fields on a T21 Gaussian grid, the same box can be selected with selindexbox
by:
cdo s e l i n d e x b o x , 2 3 , 4 8 , 1 3 , 2 0 i f i l e

ofile

35

Conditional selection

Reference manual

2.4. Conditional selection

This section contains modules to conditional select field elements. The fields in the first input file are
handled as a mask. A value not equal to zero is treated as true, zero is treated as false.
Here is a short overview of all operators in this section:

36

ifthen
ifnotthen

If then
If not then

ifthenelse

If then else

ifthenc
ifnotthenc

If then constant
If not then constant

Reference manual

Conditional selection

2.4.1. COND - Conditional select one field

Synopsis
< operator > ifile1 ifile2 ofile

Description
This module selects field elements from ifile2 with respect to ifile1 and writes them to ofile.
The fields in ifile1 are handled as a mask. A value not equal to zero is treated as true, zero is
treated as false. The number of fields in ifile1 has either to be the same as in ifile2 or the same
as in one time step of ifile2 or only one. The fields in ofile inherit the meta data from ifile2.

Operators
ifthen

If then

o(t, x) =

ifnotthen

i2 (t, x) if i1 ([t, ]x) 6= 0 i1 ([t, ]x) 6= miss

miss
if i1 ([t, ]x) = 0 i1 ([t, ]x) = miss

If not then

i2 (t, x) if i1 ([t, ]x) = 0 i1 ([t, ]x) 6= miss
o(t, x) =
miss
if i1 ([t, ]x) 6= 0 i1 ([t, ]x) = miss

Example
To select all field elements of ifile2 if the corresponding field element of ifile1 is greater than 0
use:
cdo i f t h e n i f i l e 1

ifile2

ofile

2.4.2. COND2 - Conditional select two fields

Synopsis
ifthenelse ifile1 ifile2 ifile3 ofile

Description
This operator selects field elements from ifile2 or ifile3 with respect to ifile1 and writes them
to ofile. The fields in ifile1 are handled as a mask. A value not equal to zero is treated as true,
zero is treated as false. The number of fields in ifile1 has either to be the same as in ifile2
or the same as in one time step of ifile2 or only one. ifile2 and ifile3 need to have the same
number of fields. The fields in ofile inherit the meta data from ifile2.

i1 ([t, ]x) 6= miss

i2 (t, x) if i1 ([t, ]x) 6= 0
o(t, x) =
i3 (t, x) if i1 ([t, ]x) = 0
i1 ([t, ]x) 6= miss

miss
if i1 ([t, ]x) = miss

Example
To select all field elements of ifile2 if the corresponding field element of ifile1 is greater than 0
and from ifile3 otherwise use:
cdo i f t h e n e l s e

ifile1

ifile2

ifile3

ofile

37

Conditional selection

Reference manual

2.4.3. CONDC - Conditional select a constant

Synopsis
< operator >,c ifile ofile

Description
This module creates fields with a constant value or missing value. The fields in ifile are handled as
a mask. A value not equal to zero is treated as true, zero is treated as false.

Operators
ifthenc

If then constant

c
if i(t, x) 6= 0 i(t, x) 6= miss
o(t, x) =
miss if i(t, x) = 0 i(t, x) = miss

ifnotthenc

If not then
 constant
c
if i(t, x) = 0 i(t, x) 6= miss
o(t, x) =
miss if i(t, x) 6= 0 i(t, x) = miss

Parameter
c

FLOAT

Constant

Example
To create fields with the constant value 7 if the corresponding field element of ifile is greater than
0 use:
cdo i f t h e n c , 7

38

ifile

ofile

Reference manual

Comparison

2.5. Comparison
This section contains modules to compare datasets. The resulting field is a mask containing 1 if the
comparison is true and 0 if not.
Here is a short overview of all operators in this section:
eq
ne
le
lt
ge
gt

Equal
Not equal
Less equal
Less than
Greater equal
Greater than

eqc
nec
lec
ltc
gec
gtc

Equal constant
Not equal constant
Less equal constant
Less than constant
Greater equal constant
Greater than constant

39

Comparison

Reference manual

2.5.1. COMP - Comparison of two fields

Synopsis
< operator > ifile1 ifile2 ofile

Description
This module compares two datasets field by field. The resulting field is a mask containing 1 if the
comparison is true and 0 if not. The number of fields in ifile1 should be the same as in ifile2.
One of the input files can contain only one time step or one field. The fields in ofile inherit the meta
data from ifile1 or ifile2. The type of comparison depends on the chosen operator.

Operators
eq

Equal
o(t, x) =

ne

le

lt

1
if i1 (t, x) = i2 (t, x) i1 (t, x), i2 (t, x) 6= miss
0
if i1 (t, x) 6= i2 (t, x) i1 (t, x), i2 (t, x) 6= miss

miss if i1 (t, x) = miss

i2 (t, x) = miss

Not equal

o(t, x) =

1
if i1 (t, x) 6= i2 (t, x) i1 (t, x), i2 (t, x) 6= miss
0
if i1 (t, x) = i2 (t, x) i1 (t, x), i2 (t, x) 6= miss
miss if i1 (t, x) = miss
i2 (t, x) = miss

Less equal

o(t, x) =

1
if i1 (t, x) i2 (t, x) i1 (t, x), i2 (t, x) 6= miss
0
if i1 (t, x) > i2 (t, x) i1 (t, x), i2 (t, x) 6= miss
miss if i1 (t, x) = miss
i2 (t, x) = miss

Less than

o(t, x) =

1
if i1 (t, x) < i2 (t, x) i1 (t, x), i2 (t, x) 6= miss
0
if i1 (t, x) i2 (t, x) i1 (t, x), i2 (t, x) 6= miss
miss if i1 (t, x) = miss
i2 (t, x) = miss

ge

Greater equal

if i1 (t, x) i2 (t, x) i1 (t, x), i2 (t, x) 6= miss

1
o(t, x) =
0
if i1 (t, x) < i2 (t, x) i1 (t, x), i2 (t, x) 6= miss

miss if i1 (t, x) = miss

i2 (t, x) = miss

gt

Greater than

if i1 (t, x) > i2 (t, x) i1 (t, x), i2 (t, x) 6= miss

1
o(t, x) =
0
if i1 (t, x) i2 (t, x) i1 (t, x), i2 (t, x) 6= miss

miss if i1 (t, x) = miss

i2 (t, x) = miss

Example
To create a mask containing 1 if the elements of two fields are the same and 0 if the elements are
different use:
cdo eq i f i l e 1

40

ifile2

ofile

Reference manual

Comparison

2.5.2. COMPC - Comparison of a field with a constant

Synopsis
< operator >,c ifile ofile

Description
This module compares all fields of a dataset with a constant. The resulting field is a mask containing
1 if the comparison is true and 0 if not. The type of comparison depends on the chosen operator.

Operators
eqc

Equal constant

1
o(t, x) =
0

miss

nec

lec

ltc

gec

gtc

if i(t, x) = c
if i(t, x) 6= c
if i(t, x) = miss

i(t, x), c 6= miss

i(t, x), c 6= miss
c = miss

Not equalconstant
if i(t, x) 6= c
1
o(t, x) =
0
if i(t, x) = c

miss if i(t, x) = miss

i(t, x), c 6= miss

i(t, x), c 6= miss
c = miss

Less equal
constant
if i(t, x) c
1
o(t, x) =
0
if i(t, x) > c

miss if i(t, x) = miss

i(t, x), c 6= miss

i(t, x), c 6= miss
c = miss

Less thanconstant
if i(t, x) < c
1
o(t, x) =
0
if i(t, x) c

miss if i(t, x) = miss

i(t, x), c 6= miss

i(t, x), c 6= miss
c = miss

Greater equal
constant
if i(t, x) c
1
o(t, x) =
0
if i(t, x) < c

miss if i(t, x) = miss

i(t, x), c 6= miss

i(t, x), c 6= miss
c = miss

Greater than
constant
if i(t, x) > c
1
o(t, x) =
0
if i(t, x) c

miss if i(t, x) = miss

i(t, x), c 6= miss

i(t, x), c 6= miss
c = miss

Parameter
c

FLOAT

Constant

Example
To create a mask containing 1 if the field element is greater than 273.15 and 0 if not use:
cdo gtc , 2 7 3 . 1 5

ifile

ofile

41

Modification

Reference manual

2.6. Modification
This section contains modules to modify the metadata, fields or part of a field in a dataset.
Here is a short overview of all operators in this section:

42

setpartab
setcode
setname
setlevel
setltype

Set
Set
Set
Set
Set

parameter table
code number
variable name
level
GRIB level type

setdate
settime
setday
setmon
setyear
settunits
settaxis
setreftime
setcalendar
shifttime

Set date
Set time of the day
Set day
Set month
Set year
Set time units
Set time axis
Set reference time
Set calendar
Shift time steps

chcode
chname
chlevel
chlevelc
chlevelv

Change
Change
Change
Change
Change

setgrid
setgridtype

Set grid
Set grid type

setzaxis

Set z-axis

setgatt
setgatts

Set global attribute

Set global attributes

invertlat

Invert latitudes

invertlev

Invert levels

maskregion

Mask regions

masklonlatbox
maskindexbox

Mask a longitude/latitude box

Mask an index box

setclonlatbox
setcindexbox

Set a longitude/latitude box to constant

Set an index box to constant

enlarge

Enlarge fields

setmissval
setctomiss
setmisstoc
setrtomiss
setvrange

Set
Set
Set
Set
Set

code number
variable name
level
level of one code
level of one variable

a new missing value

constant to missing value
missing value to constant
range to missing value
valid range

Reference manual

Modification

2.6.1. SET - Set field info

Synopsis
setpartab,table ifile ofile
setcode,code ifile ofile
setname,name ifile ofile
setlevel,level ifile ofile
setltype,ltype ifile ofile

Description
This module sets some field information. Depending on the chosen operator the parameter table,
code number, variable name or level is set.

Operators
setpartab

Set parameter table

Sets the parameter table for all variables.

setcode

Set code number

Sets the code number for all variables to the same given value.

setname

Set variable name

Sets the name of the first variable.

setlevel

Set level
Sets the first level of all variables.

setltype

Set GRIB level type

Sets the GRIB level type of all variables.

Parameter
table

STRING

Parameter table file or name

code

INTEGER

Code number

name

STRING

Variable name

level

FLOAT

New level

ltype

INTEGER

GRIB level type

Example
To assign the parameter table echam5 to the input dataset use:
cdo s e t p a r t a b , echam5 i f i l e

ofile

43

Modification

Reference manual

2.6.2. SETTIME - Set time

Synopsis
setdate,date ifile ofile
settime,time ifile ofile
setday,day ifile ofile
setmon,month ifile ofile
setyear,year ifile ofile
settunits,units ifile ofile
settaxis,date,time[,inc] ifile ofile
setreftime,date,time[,units] ifile ofile
setcalendar,calendar ifile ofile
shifttime,sval ifile ofile

Description
This module sets the time axis or part of the time axis. Which part of the time axis is overwritten
depends on the chosen operator.

Operators

44

setdate

Set date
Sets the date in every time step to the same given value.

settime

Set time of the day

Sets the time in every time step to the same given value.

setday

Set day
Sets the day in every time step to the same given value.

setmon

Set month
Sets the month in every time step to the same given value.

setyear

Set year
Sets the year in every time step to the same given value.

settunits

Set time units

Sets the base units of a relative time axis.

settaxis

Set time axis

Sets the time axis.

setreftime

Set reference time

Sets the reference time of a relative time axis.

setcalendar

Set calendar
Sets the calendar of a relative time axis.

shifttime

Shift time steps

Shifts all time steps by the parameter sval.

Reference manual

Modification

Parameter
day

INTEGER

Value of the new day

month

INTEGER

Value of the new month

year

INTEGER

Value of the new year

units

STRING

Base units of the time axis (seconds, minutes, hours, days, months, years)

date

STRING

Date (format YYYY-MM-DD)

time

STRING

Time (format HH:MM)

inc

STRING

Optional increment (seconds, minutes, hours, days, months, years) [default:

0hour]
calendar

STRING

Calendar (standard, proleptic, 360days, 365days, 366days)

sval

STRING

Shift value (e.g. -3hour)

Example
To set the time axis to 1987-01-16 12:00 with an increment of one month for each time step use:
cdo s e t t a x i s ,1987 01 16 ,12:00 ,1mon i f i l e

ofile

Result of cdo showdate ofile for a dataset with 12 time steps:

19870116 19870216 19870316 19870416 19870516 19870616 \
19870716 19870816 19870916 19871016 19871116 19871216

To shift this time axis by -15 days use:

cdo s h i f t t i m e , 15 days i f i l e

ofile

Result of cdo showdate ofile:

19870101 19870201 19870301 19870401 19870501 19870601 \
19870701 19870801 19870901 19871001 19871101 19871201

45

Modification

Reference manual

2.6.3. CHANGE - Change field header

Synopsis
chcode,oldcode,newcode[,...] ifile ofile
chname,oldname,newname,... ifile ofile
chlevel,oldlev,newlev,... ifile ofile
chlevelc,code,oldlev,newlev ifile ofile
chlevelv,name,oldlev,newlev ifile ofile

Description
This module reads fields from ifile, changes some header values and writes the results to ofile.
The kind of changes depends on the chosen operator.

Operators
chcode

Change code number

Changes some user given code numbers to new user given values.

chname

Change variable name

Changes some user given variable names to new user given names.

chlevel

Change level
Changes some user given levels to new user given values.

chlevelc

Change level of one code

Changes one level of a user given code number.

chlevelv

Change level of one variable

Changes one level of a user given variable name.

Parameter
code

INTEGER

Code number

oldcode,newcode,...

INTEGER

Pairs of old and new code numbers

name

STRING

Variable name

oldname,newname,...

STRING

Pairs of old and new variable names

oldlev

FLOAT

Old level

newlev

FLOAT

New level

oldlev,newlev,...

FLOAT

Pairs of old and new levels

Example
To change the code number 98 to 179 and 99 to 211 use:
cdo chcode , 9 8 , 1 7 9 , 9 9 , 2 1 1 i f i l e

46

ofile

Reference manual

Modification

2.6.4. SETGRID - Set grid type

Synopsis
setgrid,grid ifile ofile
setgridtype,gridtype ifile ofile

Description
This module sets the grid description of all fields with the same grid size as the new grid.

Operators
setgrid

Set grid
Sets the grid description of all fields.

setgridtype

Set grid type

Sets the grid type of all grids to a user given value.

Parameter
grid

STRING

Target grid description file or name

gridtype

STRING

Target grid type (curvilinear or cell)

Example
Assuming a dataset has fields with 2048 gridpoints without or with wrong grid description. To set
the grid description of all input fields to a T21 Gaussian grid (2048 gridpoints) use:
cdo s e t g r i d , t 2 1 g r i d i f i l e

ofile

2.6.5. SETZAXIS - Set z-axis type

Synopsis
setzaxis,zaxis ifile ofile

Description
This operator sets the z-axis description of all variables with the same number of level as the new
z-axis.

Parameter
zaxis

STRING

Z-axis description file or name of the target z-axis

47

Modification

Reference manual

2.6.6. SETGATT - Set global attribute

Synopsis
setgatt,attname,attstring ifile ofile
setgatts,attfile ifile ofile

Description
This module sets global text attributes of a dataset. Depending on the chosen operator the attributes
are read from a file or can be specified by a parameter.

Operators
setgatt

Set global attribute

Sets one user defined global text attribute.

setgatts

Set global attributes

Sets user defined global text attributes. The name and text of the global attributes are
read from a file.

Parameter
attname,attstring

STRING

Name and text of the global attribute (without spaces!)

attfile

STRING

File name which contains global text attributes

Note
Besides netCDF none of the supported data formats supports global attributes.

Example
To set the global text attribute myatt to myattcontents in a netCDF file use:
cdo s e t g a t t , myatt , m y a t t c o n t e n t s i f i l e

Result of ncdump -h ofile:

netcdf o f i l e {
dimensions : . . .
variables :

...

// g l o b a l a t t r i b u t e s :
: myatt = m y a t tc o n t e n s ;
}

48

ofile

Reference manual

Modification

2.6.7. INVERT - Invert latitudes

Synopsis
invertlat ifile ofile

Description
This operator inverts the latitudes of all fields with a regular lon/lat grid.

Example
To invert the latitudes of a 2D field from N->S to S->N use:
cdo i n v e r t l a t

ifile

ofile

2.6.8. INVERTLEV - Invert levels

Synopsis
invertlev ifile ofile

Description
This operator inverts the levels of all non hybrid 3D variables.

49

Modification

Reference manual

2.6.9. MASKREGION - Mask regions

Synopsis
maskregion,regions ifile ofile

Description
Masks different regions of fields with a regular lon/lat grid. The elements inside a region are untouched, the elements outside are set to missing value. All input fields must have the same horizontal
grid. The user has to give ASCII formatted files with different regions. A region is defined by a
polygon. Each line of a polygon description file contains the longitude and latitude of one point.
Each polygon description file can contain one or more polygons separated by a line with the character
&.

Parameter
regions

STRING

Comma separated list of ASCII formatted files with different regions

Example
To mask the region with the longitudes from 120E to 90W and latitudes from 20N to 20S on all input
fields use:
cdo maskregion , myregion i f i l e

ofile

For this example the polygon description file myregion should contain the following four coordinates:
120 20
120 20
270 20
270 20

50

Reference manual

Modification

2.6.10. MASKBOX - Mask a box

Synopsis
masklonlatbox,lon1,lon2,lat1,lat2 ifile ofile
maskindexbox,idx1,idx2,idy1,idy2 ifile ofile

Description
Masks a box of the rectangular understood field. The elements inside the box are untouched, the
elements outside are set to missing value. All input fields need to have the same horizontal grid. Use
sellonlatbox or selindexbox if only the data inside the box are needed.

Operators
masklonlatbox

Mask a longitude/latitude box

Masks a longitude/latitude box. The user has to give the longitudes and latitudes of the edges of the box.

maskindexbox

Mask an index box

Masks an index box. The user has to give the indexes of the edges of the box.
The index of the left edge can be greater then the one of the right edge.

Parameter
lon1

FLOAT

Western longitude

lon2

FLOAT

Eastern longitude

lat1

FLOAT

Southern or northern latitude

lat2

FLOAT

Northern or southern latitude

idx1

INTEGER

Index of first longitude

idx2

INTEGER

Index of last longitude

idy1

INTEGER

Index of first latitude

idy2

INTEGER

Index of last latitude

Example
To mask the region with the longitudes from 120E to 90W and latitudes from 20N to 20S on all input
fields use:
cdo mas kl onl atb ox ,120 , 90 ,20 , 20 i f i l e

ofile

If the input dataset has fields on a T21 Gaussian grid, the same box can be masked with maskindexbox
by:
cdo maskindexbox , 2 3 , 4 8 , 1 3 , 2 0 i f i l e

ofile

51

Modification

Reference manual

2.6.11. SETBOX - Set a box to constant

Synopsis
setclonlatbox,c,lon1,lon2,lat1,lat2 ifile ofile
setcindexbox,c,idx1,idx2,idy1,idy2 ifile ofile

Description
Sets a box of the rectangular understood field to a constant value. The elements outside the box are
untouched, the elements inside are set to the given constant. All input fields need to have the same
horizontal grid.

Operators
setclonlatbox

Set a longitude/latitude box to constant

Sets the values of a longitude/latitude box to a constant value. The user has to
give the longitudes and latitudes of the edges of the box.

setcindexbox

Set an index box to constant

Sets the values of an index box to a constant value. The user has to give the
indexes of the edges of the box. The index of the left edge can be greater than
the one of the right edge.

Parameter
c

FLOAT

Constant

lon1

FLOAT

Western longitude

lon2

FLOAT

Eastern longitude

lat1

FLOAT

Southern or northern latitude

lat2

FLOAT

Northern or southern latitude

idx1

INTEGER

Index of first longitude

idx2

INTEGER

Index of last longitude

idy1

INTEGER

Index of first latitude

idy2

INTEGER

Index of last latitude

Example
To set all values in the region with the longitudes from 120E to 90W and latitudes from 20N to 20S
to the constant value -1.23 use:
cdo s e t c l o n l a t b o x , 1.23 ,120 , 90 ,20 , 20 i f i l e

ofile

If the input dataset has fields on a T21 Gaussian grid, the same box can be set with setcindexbox by:
cdo s e t c i n d e x b o x , 1 . 2 3 , 2 3 , 4 8 , 1 3 , 2 0 i f i l e

52

ofile

Reference manual

Modification

2.6.12. ENLARGE - Enlarge fields

Synopsis
enlarge,grid ifile ofile

Description
Enlarge all fields of ifile to a user given grid. Normally only the last field element is used for the
enlargement. If however the input and output grid are regular lon/lat grids, a zonal or meridional
enlargement is possible. Zonal enlargement takes place, if the xsize of the input field is 1 and the
ysize of both grids are the same. For meridional enlargement the ysize have to be 1 and the xsize of
both grids should have the same size.

Parameter
grid

STRING

Target grid description file or name

Example
Assumed you want to add two datasets. The first dataset is on a T21 grid (2048 field elements) and
the second dataset is only a global mean (1 field element). Before you can add these two datasets the
second dataset have to be enlarged to the grid size of the first dataset:
cdo e n l a r g e , t 2 1 g r i d i f i l e 2 t m p f i l e
cdo add i f i l e 1 t m p f i l e o f i l e

Or shorter using operator piping:

cdo add i f i l e 1 e n l a r g e , t 2 1 g r i d i f i l e 2

ofile

53

Modification

Reference manual

2.6.13. SETMISS - Set missing value

Synopsis
setmissval,newmiss ifile ofile
setctomiss,c ifile ofile
setmisstoc,c ifile ofile
setrtomiss,rmin,rmax ifile ofile
setvrange,rmin,rmax ifile ofile

Description
This module sets part of a field to missing value or missing values to a constant value. Which part of
the field is set depends on the chosen operator.

Operators
setmissval

Set a newmissing value

newmiss if i(t, x) = miss
o(t, x) =
i(t, x)
if i(t, x) 6= miss

setctomiss

Set constant
 to missing value
miss if i(t, x) = c
o(t, x) =
i(t, x) if i(t, x) 6= c

setmisstoc

Set missing
 value to constant
c
if i(t, x) = miss
o(t, x) =
i(t, x) if i(t, x) 6= miss

setrtomiss

Set rangeto missing value

miss if i(t, x) rmin i(t, x) rmax
o(t, x) =
i(t, x) if i(t, x) < rmin i(t, x) > rmax

setvrange

Set valid 
range
miss
o(t, x) =
i(t, x)

if i(t, x) < rmin i(t, x) > rmax

if i(t, x) rmin i(t, x) rmax

Parameter

54

newmiss

FLOAT

New missing value

FLOAT

Constant

rmin

FLOAT

Lower bound

rmax

FLOAT

Upper bound

Reference manual

Modification

Example
Assume an input dataset has one field with temperatures in the range from 246 to 304 Kelvin. To set
all values below 273.15 Kelvin to missing value use:
cdo s e t r t o m i s s , 0 , 2 7 3 . 1 5

ifile

ofile

Result of cdo info ifile:

1 :
Date Time Code L e v e l
1 : 19871231 1 2 : 0 0 139
0

Size
2048

Miss :
0 :

Minimum
246.27

Mean
276.75

Maximum
303.71

Size
2048

Miss :
871 :

Minimum
273.16

Mean
287.08

Maximum
303.71

Result of cdo info ofile:

1 :
Date Time Code L e v e l
1 : 19871231 1 2 : 0 0 139
0

55

Arithmetic

Reference manual

2.7. Arithmetic
This section contains modules to arithmetically process datasets.
Here is a short overview of all operators in this section:

56

expr
exprf

Evaluate expressions
Evaluate expressions from script file

abs
int
nint
pow
sqr
sqrt
exp
ln
log10
sin
cos
tan
asin
acos
reci

Absolute value
Integer value
Nearest integer value
Power
Square
Square root
Exponential
Natural logarithm
Base 10 logarithm
Sine
Cosine
Tangent
Arc sine
Arc cosine
Reciprocal value

addc
subc
mulc
divc

Add a constant
Subtract a constant
Multiply with a constant
Divide by a constant

add
sub
mul
div
min
max
atan2

Add two fields

Subtract two fields
Multiply two fields
Divide two fields
Minimum of two fields
Maximum of two fields
Arc tangent of two fields

monadd
monsub
monmul
mondiv

Add monthly time series

Subtract monthly time series
Multiply monthly time series
Divide monthly time series

ymonadd
ymonsub
ymonmul
ymondiv

Add multi-year monthly time series

Subtract multi-year monthly time series
Multiply multi-year monthly time series
Divide multi-year monthly time series

muldpm
divdpm
muldpy
divdpy

Multiply with days per month

Divide by days per month
Multiply with days per year
Divide by days per year

Reference manual

Arithmetic

2.7.1. EXPR - Evaluate expressions

Synopsis
expr,instr ifile ofile
exprf,filename ifile ofile

Description
This module arithmetically processes every time step of the input dataset. Each individual assignment
statement have to end with a semi-colon. The basic arithmetic operations addition +, subtraction ,
multiplication , division / and exponentiation can be used. The following intrinsic functions are
available:
sqrt(x)

Square Root of x

exp(x)

Exponential of x

log(x)

Natural logarithm of x

log10(x)

Base 10 logarithm of x

sin(x)

Sine of x, where x is specified in radians

cos(x)

Cosine of x, where x is specified in radians

tan(x)

Tangent of x, where x is specified in radians

asin(x)

Arc-sine of x, where x is specified in radians

acos(x)

Arc-cosine of x, where x is specified in radians

atan(x)

Arc-tangent of x, where x is specified in radians

Operators
expr

Evaluate expressions
The processing instructions are read from the parameter.

exprf

Evaluate expressions from script file

Contrary to expr the processing instructions are read from a file.

Parameter
instr

STRING

Processing instructions (without spaces!)

filename

STRING

File with processing instructions

Example
Assume an input dataset contains at least the variables aprl, aprc and ts. To create a new variable
var1 with the sum of aprl and aprc and a variable var2 which convert the temperature ts from
Kelvin to Celsius use:
cdo expr , var1=a p r l+a p r c ; var2=t s 2 7 3 . 1 5 ; i f i l e

ofile

The same example, but the instructions are read from a file:
cdo e x p r f , myexpr i f i l e

ofile

The file myexpr contains:

var1 = a p r l + a p r c ;
var2 = t s 2 7 3 . 1 5 ;

57

Arithmetic

Reference manual

2.7.2. MATH - Mathematical functions

Synopsis
< operator > ifile ofile

Description
This module contains some standard mathematical functions. All trigonometric functions calculate
with radians.

Operators
abs

Absolute value
o(t, x) = abs(i(t, x))

int

Integer value
o(t, x) = int(i(t, x))

nint

Nearest integer value

o(t, x) = nint(i(t, x))

pow

Power
y
o(t, x) = i(t, x)

sqr

Square
2
o(t, x) = i(t, x)

sqrt

Square root
p
o(t, x) = i(t, x)

exp

Exponential
o(t, x) = ei(t,x)

ln

Natural logarithm
o(t, x) = ln(i(t, x))

log10

Base 10 logarithm
o(t, x) = log10 (i(t, x))

sin

Sine
o(t, x) = sin(i(t, x))

cos

Cosine
o(t, x) = cos(i(t, x))

tan

Tangent
o(t, x) = tan(i(t, x))

asin

Arc sine
o(t, x) = arcsin(i(t, x))

acos

Arc cosine
o(t, x) = arccos(i(t, x))

reci

Reciprocal value
o(t, x) = 1/i(t, x)

Example
To calculate the square root for all field elements use:
cdo s q r t

58

ifile

ofile

Reference manual

Arithmetic

2.7.3. ARITHC - Arithmetic with a constant

Synopsis
< operator >,c ifile ofile

Description
This module performs simple arithmetic with all field elements of a dataset and a constant. The fields
in ofile inherit the meta data from ifile.

Operators
addc

Add a constant
o(t, x) = i(t, x) + c

subc

Subtract a constant
o(t, x) = i(t, x) c

mulc

Multiply with a constant

o(t, x) = i(t, x) c

divc

Divide by a constant
o(t, x) = i(t, x)/c

Parameter
c

FLOAT

Constant

Example
To sum all input fields with the constant -273.15 use:
cdo addc , 273.15 i f i l e

ofile

59

Arithmetic

Reference manual

2.7.4. ARITH - Arithmetic on two datasets

Synopsis
< operator > ifile1 ifile2 ofile

Description
This module performs simple arithmetic of two datasets. The number of fields in ifile1 should be
the same as in ifile2. One of the input files can contain only one time step or one field. The fields
in ofile inherit the meta data from ifile1 or ifile2.

Operators
add

Add two fields

o(t, x) = i1 (t, x) + i2 (t, x)

sub

Subtract two fields

o(t, x) = i1 (t, x) i2 (t, x)

mul

Multiply two fields

o(t, x) = i1 (t, x) i2 (t, x)

div

Divide two fields

o(t, x) = i1 (t, x)/i2 (t, x)

min

Minimum of two fields

o(t, x) = min(i1 (t, x), i2 (t, x))

max

Maximum of two fields

o(t, x) = max(i1 (t, x), i2 (t, x))

atan2

Arc tangent of two fields

The atan2 operator calculates the arc tangent of two fields. The result is in radians, which
is between -PI and PI (inclusive).
o(t, x) = atan2(i1 (t, x), i2 (t, x))

Example
To sum all fields of the first input file with the corresponding fields of the second input file use:
cdo add i f i l e 1

60

ifile2

ofile

Reference manual

Arithmetic

2.7.5. MONARITH - Monthly arithmetic

Synopsis
< operator > ifile1 ifile2 ofile

Description
This module performs simple arithmetic of a time series and one time step with the same month
and year. For each field in ifile1 the corresponding field of the time step in ifile2 with the same
month and year is used. The header information in ifile1 have to be the same as in ifile2. Usually
ifile2 is generated by a call of the module MONSTAT.

Operators
monadd

Add monthly time series

Adds a time series and a monthly time series.

monsub

Subtract monthly time series

Subtracts a time series and a monthly time series.

monmul

Multiply monthly time series

Multiplies a time series and a monthly time series.

mondiv

Divide monthly time series

Divides a time series and a monthly time series.

Example
To subtract a monthly time average from a time series use:
cdo monsub i f i l e monavg i f i l e

ofile

61

Arithmetic

Reference manual

2.7.6. YMONARITH - Multi-year monthly arithmetic

Synopsis
< operator > ifile1 ifile2 ofile

Description
This module performs simple arithmetic of a time series and one time step with the same month of
year. For each field in ifile1 the corresponding field of the time step in ifile2 with the same month
of year is used. The header information in ifile1 have to be the same as in ifile2. Usually ifile2
is generated by a call of the module YMONSTAT.

Operators
ymonadd

Add multi-year monthly time series

Adds a time series and a multi-year monthly time series.

ymonsub

Subtract multi-year monthly time series

Subtracts a time series and a multi-year monthly time series.

ymonmul

Multiply multi-year monthly time series

Multiplies a time series and a multi-year monthly time series.

ymondiv

Divide multi-year monthly time series

Divides a time series and a multi-year monthly time series.

Example
To subtract a multi-year monthly time average from a time series use:
cdo ymonsub i f i l e ymonavg i f i l e

62

ofile

Reference manual

Arithmetic

2.7.7. ARITHDAYS - Arithmetic with days

Synopsis
< operator > ifile ofile

Description
This module multiplies or divides each time step of a dataset with the corresponding days per month
or days per year.

Operators
muldpm

Multiply with days per month

o(t, x) = i(t, x) days per month

divdpm

Divide by days per month

o(t, x) = i(t, x)/days per month

muldpy

Multiply with days per year

o(t, x) = i(t, x) days per year

divdpy

Divide by days per year

o(t, x) = i(t, x)/days per year

Example
Assume an input dataset is a monthly mean time series. To compute the yearly mean from the correct
weighted monthly mean use:
cdo muldpm i f i l e t m p f i l e 1
cdo y e a r a v g t m p f i l e 1 t m p f i l e 2
cdo mulc , 1 2 divdpy t m p f i l e 2 o f i l e

Or all in one command line:

cdo mulc , 1 2 divdpy y e a r a v g muldpm i f i l e

ofile

63

Statistical values

Reference manual

2.8. Statistical values

This section contains modules to compute statistical values of datasets. In this program there is the
different notion of mean and average to distinguish two different kinds of treatment of missing values.
While computing the mean, only the not missing values are considered to belong to the sample with the
side effect of a probably reduced sample size. Computing the average is just adding the sample members
and divide the result by the sample size. For example, the mean of 1, 2, miss and 3 is (1+2+3)/3 = 2,
whereas the average is (1+2+miss+3)/4 = miss/4 = miss. If there are no missing values in the sample,
the average and the mean are identical.
In this section the abbreviations as in the following table are used:

sum

n
X

xi

i=1

mean resp. avg

n
X

xi

i=1

mean resp. avg

weighted by
{wi , i = 1, ..., n}

Variance
var

n1

Standard deviation
std

std weighted by
{wi , i = 1, ..., n}

1
wj

j=1

n
X

wi xi

i=1

n
X

(xi x)2

i=1

var weighted by
{wi , i = 1, ..., n}

n
X

n
X

1
wj

j=1

n
X

wi xi

i=1

64

n
X

1
wj

j=1

n
X

wj xj

j=1

v
u
n
X
u
tn1
(xi x)2
i=1

2
u
1

1
u n
n
n
n
u X
X
X
X

u
wj
wi xi
wj
wj xj
t
j=1

i=1

Here is a short overview of all operators in this section:

ensmin
ensmax
enssum
ensmean
ensavg
ensvar
ensstd
enspctl

Ensemble
Ensemble
Ensemble
Ensemble
Ensemble
Ensemble
Ensemble
Ensemble

minimum
maximum
sum
mean
average
variance
standard deviation
percentiles

j=1

j=1

Reference manual

Statistical values

fldmin
fldmax
fldsum
fldmean
fldavg
fldvar
fldstd
fldpctl

Field
Field
Field
Field
Field
Field
Field
Field

minimum
maximum
sum
mean
average
variance
standard deviation
percentiles

zonmin
zonmax
zonsum
zonmean
zonavg
zonvar
zonstd
zonpctl

Zonal
Zonal
Zonal
Zonal
Zonal
Zonal
Zonal
Zonal

minimum
maximum
sum
mean
average
variance
standard deviation
percentiles

mermin
mermax
mersum
mermean
meravg
mervar
merstd
merpctl

Meridional
Meridional
Meridional
Meridional
Meridional
Meridional
Meridional
Meridional

vertmin
vertmax
vertsum
vertmean
vertavg
vertvar
vertstd

Vertical
Vertical
Vertical
Vertical
Vertical
Vertical
Vertical

timselmin
timselmax
timselsum
timselmean
timselavg
timselvar
timselstd

Time
Time
Time
Time
Time
Time
Time

timselpctl

Time range percentiles

runmin
runmax
runsum
runmean
runavg
runvar
runstd

Running
Running
Running
Running
Running
Running
Running

runpctl

Running percentiles

minimum
maximum
sum
mean
average
variance
standard deviation
percentiles

minimum
maximum
sum
mean
average
variance
standard deviation

range
range
range
range
range
range
range

minimum
maximum
sum
mean
average
variance
standard deviation

minimum
maximum
sum
mean
average
variance
standard deviation

65

Statistical values

66

Reference manual

timmin
timmax
timsum
timmean
timavg
timvar
timstd

Time
Time
Time
Time
Time
Time
Time

minimum
maximum
sum
mean
average
variance
standard deviation

timpctl

Time percentiles

hourmin
hourmax
hoursum
hourmean
houravg
hourvar
hourstd

Hourly
Hourly
Hourly
Hourly
Hourly
Hourly
Hourly

hourpctl

Hourly percentiles

daymin
daymax
daysum
daymean
dayavg
dayvar
daystd

Daily
Daily
Daily
Daily
Daily
Daily
Daily

daypctl

Daily percentiles

monmin
monmax
monsum
monmean
monavg
monvar
monstd

Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly

monpctl

Monthly percentiles

yearmin
yearmax
yearsum
yearmean
yearavg
yearvar
yearstd

Yearly
Yearly
Yearly
Yearly
Yearly
Yearly
Yearly

yearpctl

Yearly percentiles

seasmin
seasmax
seassum
seasmean
seasavg
seasvar
seasstd

Seasonal
Seasonal
Seasonal
Seasonal
Seasonal
Seasonal
Seasonal

minimum
maximum
sum
mean
average
variance
standard deviation

minimum
maximum
sum
mean
average
variance
standard deviation

minimum
maximum
sum
mean
average
variance
standard deviation

minimum
maximum
sum
mean
average
variance
standard deviation

minimum
maximum
sum
mean
average
variance
standard deviation

Reference manual

Statistical values

seaspctl

Seasonal percentiles

yhourmin
yhourmax
yhoursum
yhourmean
yhouravg
yhourvar
yhourstd

Multi-year
Multi-year
Multi-year
Multi-year
Multi-year
Multi-year
Multi-year

hourly
hourly
hourly
hourly
hourly
hourly
hourly

ydaymin
ydaymax
ydaysum
ydaymean
ydayavg
ydayvar
ydaystd

Multi-year
Multi-year
Multi-year
Multi-year
Multi-year
Multi-year
Multi-year

daily
daily
daily
daily
daily
daily
daily

ydaypctl

Multi-year daily percentiles

ymonmin
ymonmax
ymonsum
ymonmean
ymonavg
ymonvar
ymonstd

Multi-year
Multi-year
Multi-year
Multi-year
Multi-year
Multi-year
Multi-year

ymonpctl

Multi-year monthly percentiles

yseasmin
yseasmax
yseassum
yseasmean
yseasavg
yseasvar
yseasstd

Multi-year
Multi-year
Multi-year
Multi-year
Multi-year
Multi-year
Multi-year

yseaspctl

Multi-year seasonal percentiles

ydrunmin
ydrunmax
ydrunsum
ydrunmean
ydrunavg
ydrunvar
ydrunstd

Multi-year
Multi-year
Multi-year
Multi-year
Multi-year
Multi-year
Multi-year

ydrunpctl

Multi-year daily running percentiles

minimum
maximum
sum
mean
average
variance
standard deviation

minimum
maximum
sum
mean
average
variance
standard deviation

monthly
monthly
monthly
monthly
monthly
monthly
monthly

seasonal
seasonal
seasonal
seasonal
seasonal
seasonal
seasonal

daily
daily
daily
daily
daily
daily
daily

minimum
maximum
sum
mean
average
variance
standard deviation

minimum
maximum
sum
mean
average
variance
standard deviation

running
running
running
running
running
running
running

minimum
maximum
sum
mean
average
variance
standard deviation

67

Statistical values

Reference manual

2.8.1. ENSSTAT - Statistical values over an ensemble

Synopsis
ensmin ifiles ofile
ensmax ifiles ofile
enssum ifiles ofile
ensmean ifiles ofile
ensavg ifiles ofile
ensvar ifiles ofile
ensstd ifiles ofile
enspctl,p ifiles ofile

Description
This module computes statistical values over an ensemble of input files. Depending on the chosen
operator the minimum, maximum, sum, average, variance, standard deviation, or a certain percentile
over all input files is written to ofile. The date information of a time step in ofile is the date of
the first input file.

Operators
ensmin

Ensemble minimum
o(t, x) = min{i1 (t, x), i2 (t, x), , in (t, x)}

ensmax

Ensemble maximum
o(t, x) = max{i1 (t, x), i2 (t, x), , in (t, x)}

enssum

Ensemble sum
o(t, x) = sum{i1 (t, x), i2 (t, x), , in (t, x)}

ensmean

Ensemble mean
o(t, x) = mean{i1 (t, x), i2 (t, x), , in (t, x)}

ensavg

Ensemble average
o(t, x) = avg{i1 (t, x), i2 (t, x), , in (t, x)}

ensvar

Ensemble variance
o(t, x) = var{i1 (t, x), i2 (t, x), , in (t, x)}

ensstd

Ensemble standard deviation

o(t, x) = std{i1 (t, x), i2 (t, x), , in (t, x)}

enspctl

Ensemble percentiles
o(t, x) = pth percentile{i1 (t, x), i2 (t, x), , in (t, x)}

Parameter
p

68

INTEGER

Percentile number in 1, ..., 99

Reference manual

Statistical values

Example
To compute the ensemble mean over 6 input files use:
cdo ensmean i f i l e 1

ifile2

ifile3

ifile4

ifile5

ifile6

ofile

Or shorter with filename substitution:

cdo ensmean i f i l e [1 6] o f i l e

To compute the 50th percentile (median) over 6 input files use:

cdo e n s p c t l , 5 0 i f i l e 1

ifile2

ifile3

ifile4

ifile5

ifile6

ofile

69

Statistical values

Reference manual

2.8.2. FLDSTAT - Statistical values over a field

Synopsis
fldmin ifile ofile
fldmax ifile ofile
fldsum ifile ofile
fldmean ifile ofile
fldavg ifile ofile
fldvar ifile ofile
fldstd ifile ofile
fldpctl,p ifile ofile

Description
This module computes statistical values of the input fields. According to the chosen operator the field
minimum, maximum, sum, average, variance, standard deviation or a certain percentile is written to
ofile.

Operators
fldmin

Field minimum
For every gridpoint x1 , ..., xn of the same field it is:
o(t, 1) = min{i(t, x0 ), x1 < x0 xn }

fldmax

Field maximum
For every gridpoint x1 , ..., xn of the same field it is:
o(t, 1) = max{i(t, x0 ), x1 < x0 xn }

fldsum

Field sum
For every gridpoint x1 , ..., xn of the same field it is:
o(t, 1) = sum{i(t, x0 ), x1 < x0 xn }

fldmean

Field mean
For every gridpoint x1 , ..., xn of the same field it is:
o(t, 1) = mean{i(t, x0 ), x1 < x0 xn }
weighted by area weights obtained by the input field.

fldavg

Field average
For every gridpoint x1 , ..., xn of the same field it is:
o(t, 1) = avg{i(t, x0 ), x1 < x0 xn }
weighted by area weights obtained by the input field.

fldvar

Field variance
For every gridpoint x1 , ..., xn of the same field it is:
o(t, 1) = var{i(t, x0 ), x1 < x0 xn }
weighted by area weights obtained by the input field.

fldstd

Field standard deviation

For every gridpoint x1 , ..., xn of the same field it is:
o(t, 1) = std{i(t, x0 ), x1 < x0 xn }
weighted by area weights obtained by the input field.

fldpctl

70

Field percentiles
For every gridpoint x1 , ..., xn of the same field it is:
o(t, 1) = pth percentile{i(t, x0 ), x1 < x0 xn }

Reference manual

Statistical values

Parameter
p

INTEGER

Percentile number in 1, ..., 99

Example
To compute the field mean of all input fields use:
cdo fldmean i f i l e

ofile

To compute the 90th percentile of all input fields use:

cdo f l d p c t l , 9 0 i f i l e

ofile

71

Statistical values

Reference manual

2.8.3. ZONSTAT - Zonal statistical values

Synopsis
zonmin ifile ofile
zonmax ifile ofile
zonsum ifile ofile
zonmean ifile ofile
zonavg ifile ofile
zonvar ifile ofile
zonstd ifile ofile
zonpctl,p ifile ofile

Description
This module computes zonal statistical values of the input fields. According to the chosen operator
the zonal minimum, maximum, sum, average, variance, standard deviation or a certain percentile is
written to ofile. All input fields need to have the same regular quadrilateral grid.

Operators
zonmin

Zonal minimum
For every latitude the minimum over all longitudes is computed.

zonmax

Zonal maximum
For every latitude the maximum over all longitudes is computed.

zonsum

Zonal sum
For every latitude the sum over all longitudes is computed.

zonmean

Zonal mean
For every latitude the mean over all longitudes is computed.

zonavg

Zonal average
For every latitude the average over all longitudes is computed.

zonvar

Zonal variance
For every latitude the variance over all longitudes is computed.

zonstd

Zonal standard deviation

For every latitude the standard deviation over all longitudes is computed.

zonpctl

Zonal percentiles
For every latitude the pth percentile over all longitudes is computed.

Parameter
p

72

INTEGER

Percentile number in 1, ..., 99

Reference manual

Statistical values

Example
To compute the zonal mean of all input fields use:
cdo zonmean i f i l e

ofile

To compute the 50th meridional percentile (median) of all input fields use:
cdo z o n p c t l , 5 0 i f i l e

ofile

73

Statistical values

Reference manual

2.8.4. MERSTAT - Meridional statistical values

Synopsis
mermin ifile ofile
mermax ifile ofile
mersum ifile ofile
mermean ifile ofile
meravg ifile ofile
mervar ifile ofile
merstd ifile ofile
merpctl,p ifile ofile

Description
This module computes meridional statistical values of the input fields. According to the chosen
operator the meridional minimum, maximum, sum, average, variance, standard deviation or a certain
percentile is written to ofile. All input fields need to have the same regular lon/lat grid.

Operators
mermin

Meridional minimum
For every longitude the minimum over all latitudes is computed.

mermax

Meridional maximum
For every longitude the maximum over all latitudes is computed.

mersum

Meridional sum
For every longitude the sum over all latitudes is computed.

mermean

Meridional mean
For every longitude the area weighted mean over all latitudes is computed.

meravg

Meridional average
For every longitude the area weighted average over all latitudes is computed.

mervar

Meridional variance
For every longitude the variance over all latitudes is computed.

merstd

Meridional standard deviation

For every longitude the standard deviation over all latitudes is computed.

merpctl

Meridional percentiles
For every longitude the pth percentile over all latitudes is computed.

Parameter
p

74

INTEGER

Percentile number in 1, ..., 99

Reference manual

Statistical values

Example
To compute the meridional mean of all input fields use:
cdo mermean i f i l e

ofile

To compute the 50th meridional percentile (median) of all input fields use:
cdo merpctl , 5 0 i f i l e

ofile

75

Statistical values

Reference manual

2.8.5. VERTSTAT - Vertical statistical values

Synopsis
< operator > ifile ofile

Description
This module computes statistical values over all levels of the input variables. According to chosen
operator the vertical minimum, maximum, sum, average, variance or standard deviation is written to
ofile.

Operators
vertmin

Vertical minimum
For every gridpoint the minimum over all levels is computed.

vertmax

Vertical maximum
For every gridpoint the maximum over all levels is computed.

vertsum

Vertical sum
For every gridpoint the sum over all levels is computed.

vertmean

Vertical mean
For every gridpoint the mean over all levels is computed.

vertavg

Vertical average
For every gridpoint the average over all levels is computed.

vertvar

Vertical variance
For every gridpoint the variance over all levels is computed.

vertstd

Vertical standard deviation

For every gridpoint the standard deviation over all levels is computed.

Example
To compute the vertical sum of all input variables use:
cdo vertsum i f i l e

76

ofile

Reference manual

Statistical values

2.8.6. TIMSELSTAT - Time range statistical values

Synopsis
< operator >,nsets[,noffset[,nskip]] ifile ofile

Description
This module computes statistical values for a selected number of time steps. According to the chosen
operator the minimum, maximum, sum, average, variance or standard deviation of the selected time
steps is written to ofile. The date information of a time step in ofile is the date of the last
contributing time step in ifile.

Operators
timselmin

Time range minimum

For every adjacent sequence t1 , ..., tn of time steps of the same selected time range
it is:
o(t, x) = min{i(t0 , x), t1 < t0 tn }

timselmax

Time range maximum

For every adjacent sequence t1 , ..., tn of time steps of the same selected time range
it is:
o(t, x) = max{i(t0 , x), t1 < t0 tn }

timselsum

Time range sum

For every adjacent sequence t1 , ..., tn of time steps of the same selected time range
it is:
o(t, x) = sum{i(t0 , x), t1 < t0 tn }

timselmean

Time range mean

For every adjacent sequence t1 , ..., tn of time steps of the same selected time range
it is:
o(t, x) = mean{i(t0 , x), t1 < t0 tn }

timselavg

Time range average

For every adjacent sequence t1 , ..., tn of time steps of the same selected time range
it is:
o(t, x) = avg{i(t0 , x), t1 < t0 tn }

timselvar

Time range variance

For every adjacent sequence t1 , ..., tn of time steps of the same selected time range
it is:
o(t, x) = var{i(t0 , x), t1 < t0 tn }

timselstd

Time range standard deviation

For every adjacent sequence t1 , ..., tn of time steps of the same selected time range
it is:
o(t, x) = std{i(t0 , x), t1 < t0 tn }

Parameter
nsets

INTEGER

Number of input time steps for each output time step

noffset

INTEGER

Number of input time steps skipped before the first time step range (optional)

nskip

INTEGER

Number of input time steps skipped between time step ranges (optional)

77

Statistical values

Reference manual

Example
Assume an input dataset has monthly means over several years. To compute seasonal means from
monthly means the first two month have to be skipped:
cdo timselmean , 3 , 2

ifile

ofile

2.8.7. TIMSELPCTL - Time range percentile values

Synopsis
timselpctl,p,nsets[,noffset[,nskip]] ifile1 ifile2 ifile3 ofile

Description
This operator computes percentile values over a selected number of time steps in ifile1. The algorithm uses histograms with minimum and maximum bounds given in ifile2 and ifile3, respectively.
The default number of histogram bins is 101. The default can be overridden by setting the environment variable CDO PCTL NBINS to a different value. The files ifile2 and ifile3 should be the result
of corresponding timselmin and timselmax operations, respectively. The date information of a time
step in ofile is the date of the last contributing time step in ifile.
For every adjacent sequence t1 , ..., tn of time steps of the same selected time range it is:
o(t, x) = pth percentile{i(t0 , x), t1 < t0 tn }

Parameter
p

INTEGER

Percentile number in 1, ..., 99

nsets

INTEGER

Number of input time steps for each output time step

noffset

INTEGER

Number of input time steps skipped before the first time step range (optional)

nskip

INTEGER

Number of input time steps skipped between time step ranges (optional)

Environment
CDO PCTL NBINS

78

Sets the number of histogram bins. The default number is 101.

Reference manual

Statistical values

2.8.8. RUNSTAT - Running statistical values

Synopsis
< operator >,nts ifile ofile

Description
This module computes running statistical values over a selected number of time steps. Depending
on the chosen operator the minimum, maximum, sum, average, variance or standard deviation of a
selected number of consecutive time steps read from ifile is written to ofile. The date information
in ofile is the date of the middle contributing time step in ifile.

Operators
runmin

Running minimum
o(t + (nts 1)/2, x) = min{i(t, x), i(t + 1, x), ..., i(t + nts 1, x)}

runmax

Running maximum
o(t + (nts 1)/2, x) = max{i(t, x), i(t + 1, x), ..., i(t + nts 1, x)}

runsum

Running sum
o(t + (nts 1)/2, x) = sum{i(t, x), i(t + 1, x), ..., i(t + nts 1, x)}

runmean

Running mean
o(t + (nts 1)/2, x) = mean{i(t, x), i(t + 1, x), ..., i(t + nts 1, x)}

runavg

Running average
o(t + (nts 1)/2, x) = avg{i(t, x), i(t + 1, x), ..., i(t + nts 1, x)}

runvar

Running variance
o(t + (nts 1)/2, x) = std{i(t, x), i(t + 1, x), ..., i(t + nts 1, x)}

runstd

Running standard deviation

o(t + (nts 1)/2, x) = std{i(t, x), i(t + 1, x), ..., i(t + nts 1, x)}

Parameter
nts

INTEGER

Number of time steps

Environment
RUNSTAT DATE

Sets the date information in ofile to the first, last or middle contributing
time step in ifile.

Example
To compute the running mean over 9 time steps use:
cdo runmean , 9

ifile

ofile

79

Statistical values

Reference manual

2.8.9. RUNPCTL - Running percentile values

Synopsis
runpctl,p,nts ifile1 ofile

Description
This module computes running percentiles over a selected number of time steps in ifile1. The date
information in ofile is the date of the medium contributing time step in ifile1.
o(t + (nts 1)/2, x) = pth percentile{i(t, x), i(t + 1, x), ..., i(t + nts 1, x)}

Parameter
p

INTEGER

Percentile number in 1, ..., 99

nts

INTEGER

Number of time steps

Example
To compute the running 50th percentile (median) over 9 time steps use:
cdo r u n p c t l , 5 0 , 9 i f i l e runmin , 9

80

i f i l e runmax , 9

ifile

ofile

Reference manual

Statistical values

2.8.10. TIMSTAT - Statistical values over all time steps

Synopsis
< operator > ifile ofile

Description
This module computes statistical values over all time steps in ifile. Depending on the chosen
operator the minimum, maximum, sum, average, variance or standard deviation of all time steps read
from ifile is written to ofile. The date information of a time step in ofile is the date of the last
contributing time step in ifile.

Operators
timmin

Time minimum
o(1, x) = min{i(t0 , x), t1 < t0 tn }

timmax

Time maximum
o(1, x) = max{i(t0 , x), t1 < t0 tn }

timsum

Time sum
o(1, x) = sum{i(t0 , x), t1 < t0 tn }

timmean

Time mean
o(1, x) = mean{i(t0 , x), t1 < t0 tn }

timavg

Time average
o(1, x) = avg{i(t0 , x), t1 < t0 tn }

timvar

Time variance
o(1, x) = var{i(t0 , x), t1 < t0 tn }

timstd

Time standard deviation

o(1, x) = std{i(t0 , x), t1 < t0 tn }

Example
To compute the mean over all input time steps use:
cdo timmean i f i l e

ofile

81

Statistical values

Reference manual

2.8.11. TIMPCTL - Percentile values over all time steps

Synopsis
timpctl,p ifile1 ifile2 ifile3 ofile

Description
This operator computes percentiles over all time steps in ifile1. The algorithm uses histograms
with minimum and maximum bounds given in ifile2 and ifile3, respectively. The default number of histogram bins is 101. The default can be overridden by setting the environment variable
CDO PCTL NBINS to a different value. The files ifile2 and ifile3 should be the result of corresponding timmin and timmax operations, respectively. The date information of a time step in ofile is the
date of the last contributing time step in ifile1.
o(1, x) = pth percentile{i(t0 , x), t1 < t0 tn }

Parameter
p

INTEGER

Percentile number in 1, ..., 99

Environment
CDO PCTL NBINS

Sets the number of histogram bins. The default number is 101.

Example
To compute the 90th percentile over all input time steps use:
cdo timmin i f i l e m i n f i l e
cdo timmax i f i l e m a x f i l e
cdo t i m p c t l , 9 0 i f i l e m i n f i l e m a x f i l e o f i l e

Or shorter using operator piping:

cdo t i m p c t l , 9 0 i f i l e timmin i f i l e timmax i f i l e

82

ofile

Reference manual

Statistical values

2.8.12. HOURSTAT - Hourly statistical values

Synopsis
< operator > ifile ofile

Description
This module computes statistical values over time steps of the same hour. Depending on the chosen
operator the minimum, maximum, sum, average, variance or standard deviation of time steps of the
same hour is written to ofile. The date information of a time step in ofile is the date of the last
contributing time step in ifile.

Operators
hourmin

Hourly minimum
For every adjacent sequence t1 , ..., tn of time steps of the same hour it is:
o(t, x) = min{i(t0 , x), t1 < t0 tn }

hourmax

Hourly maximum
For every adjacent sequence t1 , ..., tn of time steps of the same hour it is:
o(t, x) = max{i(t0 , x), t1 < t0 tn }

hoursum

Hourly sum
For every adjacent sequence t1 , ..., tn of time steps of the same hour it is:
o(t, x) = sum{i(t0 , x), t1 < t0 tn }

hourmean

Hourly mean
For every adjacent sequence t1 , ..., tn of time steps of the same hour it is:
o(t, x) = mean{i(t0 , x), t1 < t0 tn }

houravg

Hourly average
For every adjacent sequence t1 , ..., tn of time steps of the same hour it is:
o(t, x) = avg{i(t0 , x), t1 < t0 tn }

hourvar

Hourly variance
For every adjacent sequence t1 , ..., tn of time steps of the same hour it is:
o(t, x) = var{i(t0 , x), t1 < t0 tn }

hourstd

Hourly standard deviation

For every adjacent sequence t1 , ..., tn of time steps of the same hour it is:
o(t, x) = std{i(t0 , x), t1 < t0 tn }

Example
To compute the hourly mean of a time series use:
cdo hourmean i f i l e

ofile

83

Statistical values

Reference manual

2.8.13. HOURPCTL - Hourly percentile values

Synopsis
hourpctl,p ifile1 ifile2 ifile3 ofile

Description
This operator computes percentiles over all time steps of the same hour in ifile1. The algorithm
uses histograms with minimum and maximum bounds given in ifile2 and ifile3, respectively. The
default number of histogram bins is 101. The default can be overridden by setting the environment
variable CDO PCTL NBINS to a different value. The files ifile2 and ifile3 should be the result of
corresponding hourmin and hourmax operations, respectively. The date information of a time step in
ofile is the date of the last contributing time step in ifile1.
For every adjacent sequence t1 , ..., tn of time steps of the same hour it is:
o(t, x) = pth percentile{i(t0 , x), t1 < t0 tn }

Parameter
p

INTEGER

Percentile number in 1, ..., 99

Environment
CDO PCTL NBINS

Sets the number of histogram bins. The default number is 101.

Example
To compute the hourly 90th percentile of a time series use:
cdo hourmin i f i l e m i n f i l e
cdo hourmax i f i l e m a x f i l e
cdo h o u r p c t l , 9 0 i f i l e m i n f i l e m a x f i l e o f i l e

Or shorter using operator piping:

cdo h o u r p c t l , 9 0 i f i l e hourmin i f i l e hourmax i f i l e

84

ofile

Reference manual

Statistical values

2.8.14. DAYSTAT - Daily statistical values

Synopsis
< operator > ifile ofile

Description
This module computes statistical values over time steps of the same day. Depending on the chosen
operator the minimum, maximum, sum, average, variance or standard deviation of time steps of the
same day is written to ofile. The date information of a time step in ofile is the date of the last
contributing time step in ifile.

Operators
daymin

Daily minimum
For every adjacent sequence t1 , ..., tn of time steps of the same day it is:
o(t, x) = min{i(t0 , x), t1 < t0 tn }

daymax

Daily maximum
For every adjacent sequence t1 , ..., tn of time steps of the same day it is:
o(t, x) = max{i(t0 , x), t1 < t0 tn }

daysum

Daily sum
For every adjacent sequence t1 , ..., tn of time steps of the same day it is:
o(t, x) = sum{i(t0 , x), t1 < t0 tn }

daymean

Daily mean
For every adjacent sequence t1 , ..., tn of time steps of the same day it is:
o(t, x) = mean{i(t0 , x), t1 < t0 tn }

dayavg

Daily average
For every adjacent sequence t1 , ..., tn of time steps of the same day it is:
o(t, x) = avg{i(t0 , x), t1 < t0 tn }

dayvar

Daily variance
For every adjacent sequence t1 , ..., tn of time steps of the same day it is:
o(t, x) = var{i(t0 , x), t1 < t0 tn }

daystd

Daily standard deviation

For every adjacent sequence t1 , ..., tn of time steps of the same day it is:
o(t, x) = std{i(t0 , x), t1 < t0 tn }

Example
To compute the daily mean of a time series use:
cdo daymean i f i l e

ofile

85

Statistical values

Reference manual

2.8.15. DAYPCTL - Daily percentile values

Synopsis
daypctl,p ifile1 ifile2 ifile3 ofile

Description
This operator computes percentiles over all time steps of the same day in ifile1. The algorithm
uses histograms with minimum and maximum bounds given in ifile2 and ifile3, respectively. The
default number of histogram bins is 101. The default can be overridden by defining the environment
variable CDO PCTL NBINS. The files ifile2 and ifile3 should be the result of corresponding daymin
and daymax operations, respectively. The date information of a time step in ofile is the date of the
last contributing time step in ifile1.
For every adjacent sequence t1 , ..., tn of time steps of the same day it is:
o(t, x) = pth percentile{i(t0 , x), t1 < t0 tn }

Parameter
p

INTEGER

Percentile number in 1, ..., 99

Environment
CDO PCTL NBINS

Sets the number of histogram bins. The default number is 101.

Example
To compute the daily 90th percentile of a time series use:
cdo daymin i f i l e m i n f i l e
cdo daymax i f i l e m a x f i l e
cdo d a y p c t l , 9 0 i f i l e m i n f i l e m a x f i l e o f i l e

Or shorter using operator piping:

cdo d a y p c t l , 9 0 i f i l e daymin i f i l e daymax i f i l e

86

ofile

Reference manual

Statistical values

2.8.16. MONSTAT - Monthly statistical values

Synopsis
< operator > ifile ofile

Description
This module computes statistical values over time steps of the same month. Depending on the chosen
operator the minimum, maximum, sum, average, variance or standard deviation of time steps of the
same month is written to ofile. The date information of a time step in ofile is the date of the last
contributing time step in ifile.

Operators
monmin

Monthly minimum
For every adjacent sequence t1 , ..., tn of time steps of the same month it is:
o(t, x) = min{i(t0 , x), t1 < t0 tn }

monmax

Monthly maximum
For every adjacent sequence t1 , ..., tn of time steps of the same month it is:
o(t, x) = max{i(t0 , x), t1 < t0 tn }

monsum

Monthly sum
For every adjacent sequence t1 , ..., tn of time steps of the same month it is:
o(t, x) = sum{i(t0 , x), t1 < t0 tn }

monmean

Monthly mean
For every adjacent sequence t1 , ..., tn of time steps of the same month it is:
o(t, x) = mean{i(t0 , x), t1 < t0 tn }

monavg

Monthly average
For every adjacent sequence t1 , ..., tn of time steps of the same month it is:
o(t, x) = avg{i(t0 , x), t1 < t0 tn }

monvar

Monthly variance
For every adjacent sequence t1 , ..., tn of time steps of the same month it is:
o(t, x) = var{i(t0 , x), t1 < t0 tn }

monstd

Monthly standard deviation

For every adjacent sequence t1 , ..., tn of time steps of the same month it is:
o(t, x) = std{i(t0 , x), t1 < t0 tn }

Example
To compute the monthly mean of a time series use:
cdo monmean i f i l e

ofile

87

Statistical values

Reference manual

2.8.17. MONPCTL - Monthly percentile values

Synopsis
monpctl,p ifile1 ifile2 ifile3 ofile

Description
This operator computes percentiles over all time steps of the same month in ifile1. The algorithm
uses histograms with minimum and maximum bounds given in ifile2 and ifile3, respectively. The
default number of histogram bins is 101. The default can be overridden by setting the environment
variable CDO PCTL NBINS to a different value. The files ifile2 and ifile3 should be the result of
corresponding monmin and monmax operations, respectively. The date information of a time step in
ofile is the date of the last contributing time step in ifile1.
For every adjacent sequence t1 , ..., tn of time steps of the same month it is:
o(t, x) = pth percentile{i(t0 , x), t1 < t0 tn }

Parameter
p

INTEGER

Percentile number in 1, ..., 99

Environment
CDO PCTL NBINS

Sets the number of histogram bins. The default number is 101.

Example
To compute the monthly 90th percentile of a time series use:
cdo monmin i f i l e m i n f i l e
cdo monmax i f i l e m a x f i l e
cdo monpctl , 9 0 i f i l e m i n f i l e m a x f i l e o f i l e

Or shorter using operator piping:

cdo monpctl , 9 0 i f i l e monmin i f i l e monmax i f i l e

88

ofile

Reference manual

Statistical values

2.8.18. YEARSTAT - Yearly statistical values

Synopsis
< operator > ifile ofile

Description
This module computes statistical values over time steps of the same year. Depending on the chosen
operator the minimum, maximum, sum, average, variance or standard deviation of time steps of the
same year is written to ofile. The date information of a time step in ofile is the date of the last
contributing time step in ifile.

Operators
yearmin

Yearly minimum
For every adjacent sequence t1 , ..., tn of time steps of the same year it is:
o(t, x) = min{i(t0 , x), t1 < t0 tn }

yearmax

Yearly maximum
For every adjacent sequence t1 , ..., tn of time steps of the same year it is:
o(t, x) = max{i(t0 , x), t1 < t0 tn }

yearsum

Yearly sum
For every adjacent sequence t1 , ..., tn of time steps of the same year it is:
o(t, x) = sum{i(t0 , x), t1 < t0 tn }

yearmean

Yearly mean
For every adjacent sequence t1 , ..., tn of time steps of the same year it is:
o(t, x) = mean{i(t0 , x), t1 < t0 tn }

yearavg

Yearly average
For every adjacent sequence t1 , ..., tn of time steps of the same year it is:
o(t, x) = avg{i(t0 , x), t1 < t0 tn }

yearvar

Yearly variance
For every adjacent sequence t1 , ..., tn of time steps of the same year it is:
o(t, x) = var{i(t0 , x), t1 < t0 tn }

yearstd

Yearly standard deviation

For every adjacent sequence t1 , ..., tn of time steps of the same year it is:
o(t, x) = std{i(t0 , x), t1 < t0 tn }

Note
The operators yearmean and yearavg compute only arithmetical means!

Example
To compute the yearly mean of a time series use:
cdo yearmean i f i l e

ofile

To compute the yearly mean from the correct weighted monthly mean use:
cdo mulc , 1 2 divdpy yearmean muldpm i f i l e

ofile

89

Statistical values

Reference manual

2.8.19. YEARPCTL - Yearly percentile values

Synopsis
yearpctl,p ifile1 ifile2 ifile3 ofile

Description
This operator computes percentiles over all time steps of the same year in ifile1. The algorithm
uses histograms with minimum and maximum bounds given in ifile2 and ifile3, respectively. The
default number of histogram bins is 101. The default can be overridden by setting the environment
variable CDO PCTL NBINS to a different value. The files ifile2 and ifile3 should be the result of
corresponding yearmin and yearmax operations, respectively. The date information of a time step in
ofile is the date of the last contributing time step in ifile1.
For every adjacent sequence t1 , ..., tn of time steps of the same year it is:
o(t, x) = pth percentile{i(t0 , x), t1 < t0 tn }

Parameter
p

INTEGER

Percentile number in 1, ..., 99

Environment
CDO PCTL NBINS

Sets the number of histogram bins. The default number is 101.

Example
To compute the yearly 90th percentile of a time series use:
cdo yearmin i f i l e m i n f i l e
cdo yearmax i f i l e m a x f i l e
cdo y e a r p c t l , 9 0 i f i l e m i n f i l e m a x f i l e o f i l e

Or shorter using operator piping:

cdo y e a r p c t l , 9 0 i f i l e yearmin i f i l e yearmax i f i l e

90

ofile

Reference manual

Statistical values

2.8.20. SEASSTAT - Seasonal statistical values

Synopsis
< operator > ifile ofile

Description
This module computes statistical values over time steps of the same season. Depending on the chosen
operator the minimum, maximum, sum, average, variance or standard deviation of time steps of the
same season is written to ofile. The date information of a time step in ofile is the date of the last
contributing time step in ifile. Be careful about the first and the last output time step, they may
be incorrect values if the seasons have incomplete time steps.

Operators
seasmin

Seasonal minimum
For every adjacent sequence t1 , ..., tn of time steps of the same season it is:
o(t, x) = min{i(t0 , x), t1 < t0 tn }

seasmax

Seasonal maximum
For every adjacent sequence t1 , ..., tn of time steps of the same season it is:
o(t, x) = max{i(t0 , x), t1 < t0 tn }

seassum

Seasonal sum
For every adjacent sequence t1 , ..., tn of time steps of the same season it is:
o(t, x) = sum{i(t0 , x), t1 < t0 tn }

seasmean

Seasonal mean
For every adjacent sequence t1 , ..., tn of time steps of the same season it is:
o(t, x) = mean{i(t0 , x), t1 < t0 tn }

seasavg

Seasonal average
For every adjacent sequence t1 , ..., tn of time steps of the same season it is:
o(t, x) = avg{i(t0 , x), t1 < t0 tn }

seasvar

Seasonal variance
For every adjacent sequence t1 , ..., tn of time steps of the same season it is:
o(t, x) = var{i(t0 , x), t1 < t0 tn }

seasstd

Seasonal standard deviation

For every adjacent sequence t1 , ..., tn of time steps of the same season it is:
o(t, x) = std{i(t0 , x), t1 < t0 tn }

Example
To compute the seasonal mean of a time series use:
cdo seasmean i f i l e

ofile

91

Statistical values

Reference manual

2.8.21. SEASPCTL - Seasonal percentile values

Synopsis
seaspctl,p ifile1 ifile2 ifile3 ofile

Description
This operator computes percentiles over all time steps in ifile1 of the same season. The algorithm
uses histograms with minimum and maximum bounds given in ifile2 and ifile3, respectively. The
default number of histogram bins is 101. The default can be overridden by setting the environment
variable CDO PCTL NBINS to a different value. The files ifile2 and ifile3 should be the result of
corresponding seasmin and seasmax operations, respectively. The date information of a time step in
ofile is the date of the last contributing time step in ifile1. Be careful about the first and the last
output time step, they may be incorrect values if the seasons have incomplete time steps.
For every adjacent sequence t1 , ..., tn of time steps of the same season it is:
o(t, x) = pth percentile{i(t0 , x), t1 < t0 tn }

Parameter
p

INTEGER

Percentile number in 1, ..., 99

Environment
CDO PCTL NBINS

Sets the number of histogram bins. The default number is 101.

Example
To compute the seasonal 90th percentile of a time series use:
cdo s e a s m i n i f i l e m i n f i l e
cdo seasmax i f i l e m a x f i l e
cdo s e a s p c t l , 9 0 i f i l e m i n f i l e m a x f i l e o f i l e

Or shorter using operator piping:

cdo s e a s p c t l , 9 0 i f i l e s e a s m i n i f i l e seasmax i f i l e

92

ofile

Reference manual

Statistical values

2.8.22. YHOURSTAT - Multi-year hourly statistical values

Synopsis
< operator > ifile ofile

Description
This module computes statistical values of each hour and day of year. Depending on the chosen
operator the minimum, maximum, sum, average, variance or standard deviation of each hour and day
of year in ifile is written to ofile. The date information in an output field is the date of the last
contributing input field.

Operators
yhourmin

Multi-year hourly minimum

o(0001, x) = min{i(t, x), day(i(t)) = 0001}
..
.
o(8784, x) = min{i(t, x), day(i(t)) = 8784}

yhourmax

Multi-year hourly maximum

o(0001, x) = max{i(t, x), day(i(t)) = 0001}
..
.
o(8784, x) = max{i(t, x), day(i(t)) = 8784}

yhoursum

Multi-year hourly sum

o(0001, x) = sum{i(t, x), day(i(t)) = 0001}
..
.
o(8784, x) = sum{i(t, x), day(i(t)) = 8784}

yhourmean

Multi-year hourly mean

o(0001, x) = mean{i(t, x), day(i(t)) = 0001}
..
.

yhouravg

Multi-year hourly average

o(0001, x) = avg{i(t, x), day(i(t)) = 0001}
..
.

o(8784, x) = mean{i(t, x), day(i(t)) = 8784}

o(8784, x) = avg{i(t, x), day(i(t)) = 8784}

yhourvar

Multi-year hourly variance

o(0001, x) = var{i(t, x), day(i(t)) = 0001}
..
.
o(8784, x) = var{i(t, x), day(i(t)) = 8784}

yhourstd

Multi-year hourly standard deviation

o(0001, x) = std{i(t, x), day(i(t)) = 0001}
..
.
o(8784, x) = std{i(t, x), day(i(t)) = 8784}

Example
To compute the hourly mean for all days over all input years use:
cdo yhourmean i f i l e

ofile

93

Statistical values

Reference manual

2.8.23. YDAYSTAT - Multi-year daily statistical values

Synopsis
< operator > ifile ofile

Description
This module computes statistical values of each day of year. Depending on the chosen operator the
minimum, maximum, sum, average, variance or standard deviation of each day of year in ifile is
written to ofile. The date information in an output field is the date of the last contributing input
field.

Operators
ydaymin

Multi-year daily minimum

o(001, x) = min{i(t, x), day(i(t)) = 001}
..
.
o(366, x) = min{i(t, x), day(i(t)) = 366}

ydaymax

Multi-year daily maximum

o(001, x) = max{i(t, x), day(i(t)) = 001}
..
.
o(366, x) = max{i(t, x), day(i(t)) = 366}

ydaysum

Multi-year daily sum

o(001, x) = sum{i(t, x), day(i(t)) = 001}
..
.

ydaymean

Multi-year daily mean

o(001, x) = mean{i(t, x), day(i(t)) = 001}
..
.

o(366, x) = sum{i(t, x), day(i(t)) = 366}

o(366, x) = mean{i(t, x), day(i(t)) = 366}

ydayavg

Multi-year daily average

o(001, x) = avg{i(t, x), day(i(t)) = 001}
..
.
o(366, x) = avg{i(t, x), day(i(t)) = 366}

ydayvar

Multi-year daily variance

o(001, x) = var{i(t, x), day(i(t)) = 001}
..
.
o(366, x) = var{i(t, x), day(i(t)) = 366}

ydaystd

Multi-year daily standard deviation

o(001, x) = std{i(t, x), day(i(t)) = 001}
..
.
o(366, x) = std{i(t, x), day(i(t)) = 366}

Example
To compute the daily mean over all input years use:
cdo ydaymean i f i l e

94

ofile

Reference manual

Statistical values

2.8.24. YDAYPCTL - Multi-year daily percentile values

Synopsis
ydaypctl,p ifile1 ifile2 ifile3 ofile

Description
This operator writes a certain percentile of each day of year in ifile1 to ofile. The algorithm
uses histograms with minimum and maximum bounds given in ifile2 and ifile3, respectively. The
default number of histogram bins is 101. The default can be overridden by setting the environment
variable CDO PCTL NBINS to a different value. The files ifile2 and ifile3 should be the result of
corresponding ydaymin and ydaymax operations, respectively. The date information in an output
field is the date of the last contributing input field.
o(001, x) = pth percentile{i(t, x), day(i(t)) = 001}
..
.
o(366, x) = pth percentile{i(t, x), day(i(t)) = 366}

Parameter
p

INTEGER

Percentile number in 1, ..., 99

Environment
CDO PCTL NBINS

Sets the number of histogram bins. The default number is 101.

Example
To compute the daily 90th percentile over all input years use:
cdo ydaymin i f i l e m i n f i l e
cdo ydaymax i f i l e m a x f i l e
cdo y d a y p c t l , 9 0 i f i l e m i n f i l e m a x f i l e o f i l e

Or shorter using operator piping:

cdo y d a y p c t l , 9 0 i f i l e ydaymin i f i l e ydaymax i f i l e

ofile

95

Statistical values

Reference manual

2.8.25. YMONSTAT - Multi-year monthly statistical values

Synopsis
< operator > ifile ofile

Description
This module computes statistical values of each month of year. Depending on the chosen operator the
minimum, maximum, sum, average, variance or standard deviation of each month of year in ifile is
written to ofile. The date information in an output field is the date of the last contributing input
field.

Operators
ymonmin

Multi-year monthly minimum

o(01, x) = min{i(t, x), month(i(t)) = 01}
..
.
o(12, x) = min{i(t, x), month(i(t)) = 12}

ymonmax

Multi-year monthly maximum

o(01, x) = max{i(t, x), month(i(t)) = 01}
..
.
o(12, x) = max{i(t, x), month(i(t)) = 12}

ymonsum

Multi-year monthly sum

o(01, x) = sum{i(t, x), month(i(t)) = 01}
..
.

ymonmean

Multi-year monthly mean

o(01, x) = mean{i(t, x), month(i(t)) = 01}
..
.

o(12, x) = sum{i(t, x), month(i(t)) = 12}

o(12, x) = mean{i(t, x), month(i(t)) = 12}

ymonavg

Multi-year monthly average

o(01, x) = avg{i(t, x), month(i(t)) = 01}
..
.
o(12, x) = avg{i(t, x), month(i(t)) = 12}

ymonvar

Multi-year monthly variance

o(01, x) = var{i(t, x), month(i(t)) = 01}
..
.
o(12, x) = var{i(t, x), month(i(t)) = 12}

ymonstd

Multi-year monthly standard deviation

o(01, x) = std{i(t, x), month(i(t)) = 01}
..
.
o(12, x) = std{i(t, x), month(i(t)) = 12}

Example
To compute the monthly mean over all input years use:
cdo ymonmean i f i l e

96

ofile

Reference manual

Statistical values

2.8.26. YMONPCTL - Multi-year monthly percentile values

Synopsis
ymonpctl,p ifile1 ifile2 ifile3 ofile

Description
This operator writes a certain percentile of each month of year in ifile1 to ofile. The algorithm
uses histograms with minimum and maximum bounds given in ifile2 and ifile3, respectively. The
default number of histogram bins is 101. The default can be overridden by setting the environment
variable CDO PCTL NBINS to a different value. The files ifile2 and ifile3 should be the result of
corresponding ymonmin and ymonmax operations, respectively. The date information in an output
field is the date of the last contributing input field.
o(01, x) = pth percentile{i(t, x), month(i(t)) = 01}
..
.
o(12, x) = pth percentile{i(t, x), month(i(t)) = 12}

Parameter
p

INTEGER

Percentile number in 1, ..., 99

Environment
CDO PCTL NBINS

Sets the number of histogram bins. The default number is 101.

Example
To compute the monthly 90th percentile over all input years use:
cdo ymonmin i f i l e m i n f i l e
cdo ymonmax i f i l e m a x f i l e
cdo ymonpctl , 9 0 i f i l e m i n f i l e m a x f i l e o f i l e

Or shorter using operator piping:

cdo ymonpctl , 9 0 i f i l e ymonmin i f i l e ymonmax i f i l e

ofile

97

Statistical values

Reference manual

2.8.27. YSEASSTAT - Multi-year seasonal statistical values

Synopsis
< operator > ifile ofile

Description
This module computes statistical values of each season. Depending on the chosen operator the
minimum, maximum, sum, average, variance or standard deviation of each season in ifile is written
to ofile. The date information in an output field is the date of the last contributing input field.

Operators
yseasmin

yseasmax

yseassum

yseasmean

yseasavg

yseasvar

yseasstd

98

Multi-year seasonal minimum

o(1, x) = min{i(t, x), month(i(t)) = 12,
o(2, x) = min{i(t, x), month(i(t)) = 03,
o(3, x) = min{i(t, x), month(i(t)) = 06,
o(4, x) = min{i(t, x), month(i(t)) = 09,

01,
04,
07,
10,

02}
05}
08}
11}

Multi-year seasonal maximum

o(1, x) = max{i(t, x), month(i(t)) = 12,
o(2, x) = max{i(t, x), month(i(t)) = 03,
o(3, x) = max{i(t, x), month(i(t)) = 06,
o(4, x) = max{i(t, x), month(i(t)) = 09,

01,
04,
07,
10,

02}
05}
08}
11}

Multi-year seasonal sum

o(1, x) = sum{i(t, x), month(i(t)) = 12,
o(2, x) = sum{i(t, x), month(i(t)) = 03,
o(3, x) = sum{i(t, x), month(i(t)) = 06,
o(4, x) = sum{i(t, x), month(i(t)) = 09,

01,
04,
07,
10,

02}
05}
08}
11}

Multi-year seasonal mean

o(1, x) = mean{i(t, x), month(i(t)) = 12,
o(2, x) = mean{i(t, x), month(i(t)) = 03,
o(3, x) = mean{i(t, x), month(i(t)) = 06,
o(4, x) = mean{i(t, x), month(i(t)) = 09,

01,
04,
07,
10,

02}
05}
08}
11}

Multi-year seasonal average

o(1, x) = avg{i(t, x), month(i(t)) = 12,
o(2, x) = avg{i(t, x), month(i(t)) = 03,
o(3, x) = avg{i(t, x), month(i(t)) = 06,
o(4, x) = avg{i(t, x), month(i(t)) = 09,

01,
04,
07,
10,

02}
05}
08}
11}

Multi-year seasonal variance

o(1, x) = var{i(t, x), month(i(t)) = 12,
o(2, x) = var{i(t, x), month(i(t)) = 03,
o(3, x) = var{i(t, x), month(i(t)) = 06,
o(4, x) = var{i(t, x), month(i(t)) = 09,

01,
04,
07,
10,

02}
05}
08}
11}

Multi-year seasonal standard deviation

o(1, x) = std{i(t, x), month(i(t)) = 12,
o(2, x) = std{i(t, x), month(i(t)) = 03,
o(3, x) = std{i(t, x), month(i(t)) = 06,
o(4, x) = std{i(t, x), month(i(t)) = 09,

01,
04,
07,
10,

02}
05}
08}
11}

Reference manual

Statistical values

Example
To compute the seasonal mean over all input years use:
cdo yseasmean i f i l e

ofile

2.8.28. YSEASPCTL - Multi-year seasonal percentile values

Synopsis
yseaspctl,p ifile1 ifile2 ifile3 ofile

Description
This operator writes a certain percentile of each season in ifile1 to ofile. The algorithm uses
histograms with minimum and maximum bounds given in ifile2 and ifile3, respectively. The
default number of histogram bins is 101. The default can be overridden by setting the environment
variable CDO PCTL NBINS to a different value. The files ifile2 and ifile3 should be the result of
corresponding yseasmin and yseasmax operations, respectively. The date information in an output
field is the date of the last contributing input field.
o(1, x) = pth
o(2, x) = pth
o(3, x) = pth
o(4, x) = pth

percentile{i(t, x), month(i(t)) = 12,

percentile{i(t, x), month(i(t)) = 03,
percentile{i(t, x), month(i(t)) = 06,
percentile{i(t, x), month(i(t)) = 09,

01,
04,
07,
10,

02}
05}
08}
11}

Parameter
p

INTEGER

Percentile number in 1, ..., 99

Environment
CDO PCTL NBINS

Sets the number of histogram bins. The default number is 101.

Example
To compute the seasonal 90th percentile over all input years use:
cdo yseasmin i f i l e m i n f i l e
cdo yseasmax i f i l e m a x f i l e
cdo y s e a s p c t l , 9 0 i f i l e m i n f i l e m a x f i l e o f i l e

Or shorter using operator piping:

cdo y s e a s p c t l , 9 0 i f i l e yseasmin i f i l e yseasmax i f i l e

ofile

99

Statistical values

Reference manual

2.8.29. YDRUNSTAT - Multi-year daily running statistical values

Synopsis
< operator >,nts ifile ofile

Description
This module writes running statistical values for each day of year in ifile to ofile. Depending on
the chosen operator, the minimum, maximum, sum, average, variance or standard deviation of all
time steps in running windows of wich the medium time step corresponds to a certain day of year is
computed. The date information in an output field is the date of the medium time step in the last
contributing running window. Note that the operator have to be applied to a continuous time series
of daily measurements in order to yield physically meaningful results. Also note that the output time
series begins (nts-1)/2 time steps after the first time step of the input time series and ends (nts-1)/2
time steps before the last one. For input data which are complete but not continuous, such as time
series of daily measurements for the same month or season within different years, the operator yields
physically meaningful results only if the input time series does include the (nts-1)/2 days before and
after each period of interest.

Operators
ydrunmin

Multi-year daily running minimum

o(001, x) = min{i(t, x), i(t + 1, x), ..., i(t + nts 1, x); day[(i(t + (nts 1)/2)] = 001}
..
.
o(366, x) = min{i(t, x), i(t + 1, x), ..., i(t + nts 1, x); day[(i(t + (nts 1)/2)] = 366}

ydrunmax

Multi-year daily running maximum

o(001, x) = max{i(t, x), i(t + 1, x), ..., i(t + nts 1, x); day[(i(t + (nts 1)/2)] = 001}
..
.
o(366, x) = max{i(t, x), i(t + 1, x), ..., i(t + nts 1, x); day[(i(t + (nts 1)/2)] = 366}

ydrunsum

Multi-year daily running sum

o(001, x) = sum{i(t, x), i(t + 1, x), ..., i(t + nts 1, x); day[(i(t + (nts 1)/2)] = 001}
..
.
o(366, x) = sum{i(t, x), i(t + 1, x), ..., i(t + nts 1, x); day[(i(t + (nts 1)/2)] = 366}

ydrunmean

Multi-year daily running mean

o(001, x) = mean{i(t, x), i(t + 1, x), ..., i(t + nts 1, x); day[(i(t + (nts 1)/2)] = 001}
..
.
o(366, x) = mean{i(t, x), i(t + 1, x), ..., i(t + nts 1, x); day[(i(t + (nts 1)/2)] = 366}

ydrunavg

Multi-year daily running average

o(001, x) = avg{i(t, x), i(t + 1, x), ..., i(t + nts 1, x); day[(i(t + (nts 1)/2)] = 001}
..
.
o(366, x) = avg{i(t, x), i(t + 1, x), ..., i(t + nts 1, x); day[(i(t + (nts 1)/2)] = 366}

ydrunvar

Multi-year daily running variance

o(001, x) = var{i(t, x), i(t + 1, x), ..., i(t + nts 1, x); day[(i(t + (nts 1)/2)] = 001}
..
.

ydrunstd

Multi-year daily running standard deviation

o(001, x) = std{i(t, x), i(t + 1, x), ..., i(t + nts 1, x); day[(i(t + (nts 1)/2)] = 001}
..
.

o(366, x) = var{i(t, x), i(t + 1, x), ..., i(t + nts 1, x); day[(i(t + (nts 1)/2)] = 366}

o(366, x) = std{i(t, x), i(t + 1, x), ..., i(t + nts 1, x); day[(i(t + (nts 1)/2)] = 366}

100

Reference manual

Statistical values

Parameter
nts

INTEGER

Number of time steps

Example
Assume the input data provide a continuous time series of daily measurements. To compute the
running multi-year daily mean over all input time steps for a running window of five days use:
cdo ydrunmean , 5

ifile

ofile

Note that except for the standard deviation the results of the operators in this module are equivalent
to a composition of corresponding operators from the YDAYSTAT and RUNSTAT modules. For
instance, the above command yields the same result as:
cdo ydaymean runmean , 5

ifile

ofile

101

Statistical values

Reference manual

2.8.30. YDRUNPCTL - Multi-year daily running percentile values

Synopsis
ydrunpctl,p,nts ifile1 ifile2 ifile3 ofile

Description
This operator writes running percentile values for each day of year in ifile1 to ofile. A certain
percentile is computed for all time steps in running windows of which the medium time step corresponds to a certain day of year. The algorithm uses histograms with minimum and maximum bounds
given in ifile2 and ifile3, respectively. The default number of histogram bins is 101. The default
can be overridden by setting the environment variable CDO PCTL NBINS to a different value. The
files ifile2 and ifile3 should be the result of corresponding ydrunmin and ydrunmax operations,
respectively. The date information in an output field is the date of the medium time step in the last
contributing running window. Note that the operator have to be applied to a continuous time series
of daily measurements in order to yield physically meaningful results. Also note that the output time
series begins (nts-1)/2 time steps after the first time step of the input time series and ends (nts-1)/2
time steps before the last. For input data which are complete but not continuous, such as time series
of daily measurements for the same month or season within different years, the operator only yields
physically meaningful results if the input time series does include the (nts-1)/2 days before and after
each period of interest.
o(001, x) = pth percentile{i(t, x), i(t + 1, x), ..., i(t + nts 1, x); day[(i(t + (nts 1)/2)] = 001}
..
.
o(366, x) = pth percentile{i(t, x), i(t + 1, x), ..., i(t + nts 1, x); day[(i(t + (nts 1)/2)] = 366}

Parameter
p

INTEGER

Percentile number in 1, ..., 99

nts

INTEGER

Number of time steps

Environment
CDO PCTL NBINS

Sets the number of histogram bins. The default number is 101.

Example
Assume the input data provide a continuous time series of daily measurements. To compute the
running multi-year daily 90th percentile over all input time steps for a running window of five days
use:
cdo ydrunmin , 5 i f i l e m i n f i l e
cdo ydrunmax , 5 i f i l e m a x f i l e
cdo y d r u n p c t l , 9 0 , 5 i f i l e m i n f i l e m a x f i l e o f i l e

Or shorter using operator piping:

cdo y d r u n p c t l , 9 0 , 5 i f i l e ydrunmin i f i l e ydrunmax i f i l e

102

ofile

Reference manual

Regression

2.9. Regression
This sections contains modules for linear regression of time series.
Here is a short overview of all operators in this section:
regres

Regression

detrend

Detrend

trend

Trend

subtrend

Subtract trend

103

Regression

Reference manual

2.9.1. REGRES - Regression

Synopsis
regres ifile ofile

Description
The values of the input file ifile are assumed to be distributed as N (a + bt, 2 ) with unknown a, b
and 2 . This operator estimates the parameter b. For every field element x only those time steps t
belong to the sample S(x), which have i(t, x) 6= miss. It is
!
!
P
P
P
1
1
i(t, x) #S(x)
i(t0 , x)
t #S(x)
t0
o(1, x) =

t0 S(x)

tS(x)

t0 S(x)

!2
P

tS(x)

1
#S(x)

t0

P
t0 S(x)

2.9.2. DETREND - Detrend time series

Synopsis
detrend ifile ofile

Description
Every time series in ifile is linearly detrended. For every field element x only those time steps t
belong to the sample S(x), which have i(t, x) 6= miss. With

X
X
1
1
i(t, x) b(x)
t
a(x) =
#S(x)
#S(x)
tS(x)

tS(x)

and
!

!
P
b(x) =

i(t, x)

tS(x)

1
#S(x)

i(t , x)

t0 S(x)

1
#S(x)

t0 S(x)

!2
P

tS(x)

1
#S(x)

t0

t0 S(x)

it is
o(t, x) = i(t, x) (a(x) + b(x)t)

Note
This operator has to keep the fields of all time steps concurrently in the memory. If not enough
memory is available use the operators trend and subtrend.

Example
To detrend the data in ifile and to store the detrended data in ofile use:
cdo d e t r e n d i f i l e

104

ofile

Reference manual

Regression

2.9.3. TREND - Trend of time series

Synopsis
trend ifile ofile1 ofile2

Description
The values of the input file ifile are assumed to be distributed as N (a + bt, 2 ) with unknown a,
b and 2 . This operator estimates the parameter a and b. For every field element x only those time
steps t belong to the sample S(x), which have i(t, x) 6= miss. It is

X
X
1
1
i(t, x) b(x)
t
o1 (1, x) =
#S(x)
#S(x)
tS(x)

tS(x)

and
!
P
o2 (1, x) =

i(t, x)

tS(x)

1
#S(x)

i(t , x)

!
t

t0 S(x)

1
#S(x)

t0 S(x)

!2
P

tS(x)

1
#S(x)

t0

t0 S(x)

Thus the estimation for a is stored in ofile1 and that for b is stored in ofile2. To subtract the
trend from the data see operator subtrend.

2.9.4. SUBTREND - Subtract a trend

Synopsis
subtrend ifile1 ifile2 ifile3 ofile

Description
This operator is for subtracting a trend computed by the operator trend. It is
o(t, x) = i1 (t, x) (i2 (1, x) + i3 (1, x) t)
where t is the time steps.

Example
The typical call for detrending the data in ifile and storing the detrended data in ofile is:
cdo t r e n d i f i l e a f i l e b f i l e
cdo s u b t r e n d i f i l e a f i l e b f i l e

ofile

The result is identical to a call of the operator detrend:

cdo d e t r e n d i f i l e

ofile

105

Interpolation

Reference manual

2.10. Interpolation
This section contains modules to interpolate datasets. There are several operators to interpolate horizontal
fields to a new grid. Some of those operators can handle only 2D fields on a regular rectangular grid. Vertical
interpolation of 3D variables is possible from hybrid model levels to height or pressure levels. Interpolation
in time is possible between time steps and years.
Here is a short overview of all operators in this section:
remapbil
remapbic
remapdis
remapnn
remapcon
remapcon2
remaplaf

Bilinear interpolation
Bicubic interpolation
Distance-weighted average remapping
Nearest neighbor remapping
First order conservative remapping
Second order conservative remapping
Largest area fraction remapping

genbil
genbic
gendis
gennn
gencon
gencon2
genlaf

Generate
Generate
Generate
Generate
Generate
Generate
Generate

remap

SCRIP grid remapping

interpolate

PINGO grid interpolation

remapeta

Remap vertical hybrid level

ml2pl
ml2hl

Model to pressure level interpolation

Model to height level interpolation

intlevel

Linear level interpolation

inttime
intntime

Interpolation between time steps

Interpolation between time steps

intyear

Interpolation between two years

106

bilinear interpolation weights

bicubic interpolation weights
distance-weighted average remap weights
nearest neighbor remap weights
1st order conservative remap weights
2nd order conservative remap weights
largest area fraction remap weights

Reference manual

Interpolation

2.10.1. REMAPGRID - SCRIP grid interpolation

Synopsis
< operator >,grid ifile ofile

Description
This module contains operators to remap all input fields to a new horizontal grid. Each operator uses
a different remapping method. The interpolation is based on an adapted SCRIP library version. For
a detailed description of the remapping methods see [SCRIP].

Operators
remapbil

Bilinear interpolation
Performs a bilinear interpolation on all input fields. This interpolation method only
works on regular quadrilateral grids.

remapbic

Bicubic interpolation
Performs a bicubic interpolation on all input fields. This interpolation method only
works on regular quadrilateral grids.

remapdis

Distance-weighted average remapping

Performs a distance-weighted average remapping of the four nearest neighbor values
on all input fields.

remapnn

Nearest neighbor remapping

Performs a nearest neighbor remapping on all input fields.

remapcon

First order conservative remapping

Performs a first order conservative remapping on all input fields.

remapcon2

Second order conservative remapping

Performs a second order conservative remapping on all input fields.

remaplaf

Largest area fraction remapping

Performs a largest area fraction remapping on all input fields.

Parameter
grid

STRING

Target grid description file or name

Environment
NORMALIZE OPT

This variable is used to choose the normalization of the conservative remapping. By default NORMALIZE OPT is set to fracarea and will include
the destination area fraction in the output weights; other options are none
and destarea (for more information see [SCRIP]).

REMAP EXTRAPOLATE

This variable is used to switch the extrapolation feature on or off. By

default the extrapolation is enabled for remapdis, remapnn and for circular
grids.

Note
For this module the author has converted the original Fortran 90 SCRIP software to ANSI C99. If
there are any problems send a bug report to CDO and not to SCRIP!

107

Interpolation

Reference manual

Example
Say ifile contains fields on a regular quadrilateral grid. To remap all fields bilinear to a T42 Gaussian
grid type:
cdo remapbil , t 4 2 g r i d i f i l e

108

ofile

Reference manual

Interpolation

2.10.2. GENWEIGHTS - Generate SCRIP grid interpolation weights

Synopsis
< operator >,grid ifile ofile

Description
Interpolation between different horizontal grids can be a very time-consuming process. Especially
if the data are on an unstructured or a large grid. In this case the SCRIP interpolation process
can be split into two parts. Firstly the generation of the interpolation weights, which is the most
time-consuming part. These interpolation weights can be reused for every remapping process with
the operator remap. This method should be used only if all input fields are on the same grid and a
possibly mask (missing values) does not change. This module contains operators to generate SCRIP
interpolation weights of the first input field. Each operator is using a different interpolation method.

Operators
genbil

Generate bilinear interpolation weights

Generates bilinear interpolation weights and writes the result to a file. This interpolation method only works on regular quadrilateral grids.

genbic

Generate bicubic interpolation weights

Generates bicubic interpolation weights and writes the result to a file. This interpolation
method only works on regular quadrilateral grids.

gendis

Generate distance-weighted average remap weights

Generates distance-weighted average remapping weights of the four nearest neighbor
values and writes the result to a file.

gennn

Generate nearest neighbor remap weights

Generates nearest neighbor remapping weights and writes the result to a file.

gencon

Generate 1st order conservative remap weights

Generates first order conservative remapping weights and writes the result to a file.

gencon2

Generate 2nd order conservative remap weights

Generates second order conservative remapping weights and writes the result to a file.

genlaf

Generate largest area fraction remap weights

Generates largest area fraction remapping weights and writes the result to a file.

Parameter
grid

STRING

Target grid description file or name

Environment
NORMALIZE OPT

This variable is used to choose the normalization of the conservative interpolation. By default NORMALIZE OPT is set to fracarea and will
include the destination area fraction in the output weights; other options
are none and destarea (for more information see [SCRIP]).

REMAP EXTRAPOLATE

This variable is used to switch the extrapolation feature on or off. By

default the extrapolation is enabled for remapdis, remapnn and for circular
grids.

109

Interpolation

Reference manual

Note
For this module the author has converted the original Fortran 90 SCRIP software to ANSI C99. If
there are any problems send a bug report to CDO and not to SCRIP!

Example
Say ifile contains fields on a regular quadrilateral grid. To remap all fields bilinear to a T42 Gaussian
grid use:
cdo g e n b i l , t 4 2 g r i d i f i l e remapweights . nc
cdo remap , t 4 2 g r i d , remapweights . nc i f i l e o f i l e

2.10.3. REMAP - SCRIP grid remapping

Synopsis
remap,grid,weights ifile ofile

Description
This operator remaps all input fields to a new horizontal grid. The remap type and the interpolation
weights of one input grid are read from a netCDF file. The netCDF file with the weights should follow
the SCRIP convention. Normally these weights come from a previous call to module GENWEIGHTS
or were created by the original SCRIP package.

Parameter
grid

STRING

Target grid description file or name

weights

STRING

Interpolation weights (SCRIP netCDF file)

Environment
NORMALIZE OPT

This variable is used to choose the normalization of the conservative interpolation. By default NORMALIZE OPT is set to fracarea and will
include the destination area fraction in the output weights; other options
are none and destarea (for more information see [SCRIP]).

REMAP EXTRAPOLATE

This variable is used to switch the extrapolation feature on or off. By

default the extrapolation is enabled for remapdis, remapnn and for circular
grids.

Note
For this module the author has converted the original Fortran 90 SCRIP software to ANSI C99. If
there are any problems send a bug report to CDO and not to SCRIP!

Example
Say ifile contains fields on a regular quadrilateral grid. To remap all fields bilinear to a T42 Gaussian
grid use:
cdo g e n b i l , t 4 2 g r i d i f i l e remapweights . nc
cdo remap , t 4 2 g r i d , remapweights . nc i f i l e o f i l e

110

Reference manual

Interpolation

2.10.4. INTGRID - Grid interpolation

Synopsis
interpolate,grid ifile ofile

Description
This module contains operators to interpolate all input fields to a new grid. All interpolation methods
in this module only work on regular lon/lat grids.
This is the grid interpolation from PINGO. The base of the interpolation is an underlying continuous
field which is constructed in the following way:
For two neighboured longitudes x1 and x2 and two neighboured latitudes y1 and y2 of the input grid
every point at longitude x and latidude y with x1 x x2 and y1 y y2 is assigned the value
y y1
x x1
+ (a12 a11 )
x2 x1
y2 y1
(x x1 )(y y1 )
+ (a22 a21 a12 + a11 )
(x2 x1 )(y2 y1 )

a = a11 + (a21 a11 )

where aij is the value at longitude xi and latitude yj . If one out of the four values a11 , a12 , a21 , a22
is the missing value then a is also the missing value. Afterwards the underlying continuous field is
expanded by a half mesh width. For a detailed description of this interpolation method see [PINGO].

Parameter
grid

STRING

Target grid description file or name

Example
Say ifile contains fields on a regular lon/lat grid. To interpolate all fields bilinear to a T42 Gaussian
grid use:
cdo i n t g r i d b i l , t 4 2 g r i d i f i l e

ofile

111

Interpolation

Reference manual

2.10.5. REMAPETA - Remap vertical hybrid level

Synopsis
remapeta,vct[,oro] ifile ofile

Description
This operator interpolates between different vertical hybrid levels. This include the preparation of
consistent data for the free atmosphere. The procedure for the vertical interpolation is based on the
HIRLAM scheme and was adapted from [INTERA]. The vertical interpolation is based on the vertical
integration of the hydrostatic equation with few adjustments. The basic tasks are the following one:
at first integration of hydrostatic equation
extrapolation of surface pressure
Planetary Boundary-Layer (PBL) profile interpolation
interpolation in free atmosphere
merging of both profiles
final surface pressure correction

The vertical interpolation corrects the surface pressure. This is simply a cut-off or an addition of
air mass. This mass correction should not influence the geostrophic velocity field in the middle
troposhere. Therefore the total mass above a given reference level is conserved. As reference level the
geopotential height of the 500 hPa level is used. Near the surface the correction can affect the vertical
structure of the PBL. Therefore the interpolation is done using the potential temperature. But in
the free atmosphere above a certain n (n=0.8 defining the top of the PBL) the interpolation is done
linearly. After the interpolation both profiles are merged. With the resulting temperature/pressure
correction the hydrostatic equation is integrated again and adjusted to the reference level finding the
final surface pressure correction. A more detailed description of the interpolation can be found in
[INTERA]. All input fields have to be on the same horizontal grid.

Parameter
vct

STRING

File name of an ASCII dataset with the vertical coordinate table

oro

STRING

File name with the orography (surf. geopotential) of the target dataset (optional)

Note
The code numbers or the variable names of the required parameter have to follow the [ECHAM]
convention. The netCDF CF convention for vertical hybrid coordinates is not supported, yet!

Example
To remap between different hybrid model level data use:
cdo remapeta , v c t i f i l e

112

ofile

Reference manual

Interpolation

2.10.6. INTVERT - Vertical interpolation

Synopsis
ml2pl,plevels ifile ofile
ml2hl,hlevels ifile ofile

Description
Interpolate 3D variables on hybrid model levels to pressure or height levels. The input file should
contain the log. surface pressure or the surface pressure. To interpolate the temperature, the orography (surface geopotential) is also needed. The pressure, temperature, and orography are identified by
their code numbers. Supported parameter tables are: WMO standard table number 2 and ECMWF
local table number 128. Use the alias ml2plx/ml2hlx or the environment variable EXTRAPOLATE to
extrapolate missing values. All input fields have to be on the same horizontal grid.

Operators
ml2pl

Model to pressure level interpolation

Interpolates 3D variables on hybrid model levels to pressure levels.

ml2hl

Model to height level interpolation

Interpolates 3D variables on hybrid model levels to height levels. The procedure is the
same as for the operator mh2pl except for the pressure levels being calculated from the
heights by: plevel = 101325 exp(hlevel/ 7000)

Parameter
plevels

FLOAT

Pressure levels in pascal

hlevels

FLOAT

Height levels in meter (max level: 65535 m)

Environment
EXTRAPOLATE

If set to 1 extrapolate missing values.

Note
The netCDF CF convention for vertical hybrid coordinates is not supported, yet!

Example
To interpolate hybrid model level data to pressure levels of 925, 850, 500 and 200 hPa use:
cdo ml2pl , 9 2 5 0 0 , 8 5 0 0 0 , 5 0 0 0 0 , 2 0 0 0 0 i f i l e

ofile

113

Interpolation

Reference manual

2.10.7. INTLEVEL - Linear level interpolation

Synopsis
intlevel,levels ifile ofile

Description
This operator performs a linear vertical interpolation of non hybrid 3D variables.

Parameter
levels

FLOAT

Target levels

Example
To interpolate 3D variables on height levels to a new set of height levels use:
cdo i n t l e v e l , 1 0 , 5 0 , 1 0 0 , 5 0 0 , 1 0 0 0 i f i l e

114

ofile

Reference manual

Interpolation

2.10.8. INTTIME - Time interpolation

Synopsis
inttime,date,time[,inc] ifile ofile
intntime,n ifile ofile

Description
This module performs linear interpolation between time steps.

Operators
inttime

Interpolation between time steps

This operator creates a new dataset by linear interpolation between time steps. The
user has to define the start date/time with an optional increment.

intntime

Interpolation between time steps

This operator performs linear interpolation between time steps. The user has to define
the number of time steps from one time step to the next.

Parameter
date

STRING

Start date (format YYYY-MM-DD)

time

STRING

Start time (format hh:mm:ss)

inc

STRING

Optional increment (seconds, minutes, hours, days, months, years) [default:

0hour]
n

INTEGER

Number of time steps from one time step to the next

Example
Assumed a 6 hourly dataset starts at 1987-01-01 12:00. To interpolate this time series to a one hourly
dataset use:
cdo i n t t i m e ,1987 01 01 ,12:00 ,1 hour i f i l e

ofile

115

Interpolation

Reference manual

2.10.9. INTYEAR - Year interpolation

Synopsis
intyear,years ifile1 ifile2 oprefix

Description
This operator performs linear interpolation between two years time step by time step. Appends four
digits with the year to oprefix to form the output file names.

Parameter
years

INTEGER

Comma separated list of years

Example
Assume there are two monthly mean datasets over a year. The first dataset has 12 time steps for
the year 1985 and the second one for the year 1990. To interpolate the years between 1985 and 1990
month by month use:
cdo i n t y e a r , 1 9 8 6 , 1 9 8 7 , 1 9 8 8 , 1 9 8 9 i f i l e 1

i f i l e 2 year

Example result of dir year* for netCDF datasets:

y e a r 1 9 8 6 . nc y e a r 1 9 8 7 . nc y e a r 1 9 8 8 . nc y e a r 1 9 8 9 . nc

116

Reference manual

Transformation

2.11. Transformation
This section contains modules to perform spectral transformations.
Here is a short overview of all operators in this section:
sp2gp
sp2gpl
gp2sp
gp2spl
sp2sp
spcut

Spectral to gridpoint
Spectral to gridpoint (linear)
Gridpoint to spectral
Gridpoint to spectral (linear)
Spectral to spectral
Cut spectral wave number

dv2uv
dv2uvl
uv2dv
uv2dvl

Divergence and vorticity to U and V wind

Divergence and vorticity to U and V wind (linear)
U and V wind to divergence and vorticity
U and V wind to divergence and vorticity (linear)

117

Transformation

Reference manual

2.11.1. SPECTRAL - Spectral transformation

Synopsis
sp2gp ifile ofile
sp2gpl ifile ofile
gp2sp ifile ofile
gp2spl ifile ofile
sp2sp,trunc ifile ofile
spcut,wnums ifile ofile

Description
This module transforms fields on Gaussian grids to spectral coefficients and vice versa.

Operators
sp2gp

Spectral to gridpoint
Convert all fields with spectral coefficients to a regular Gaussian grid. The number of
latitudes of the resulting Gaussian grid is calculated from the triangular truncation by:
nlat = N IN T ((trunc 3 + 1.)/2.)

sp2gpl

Spectral to gridpoint (linear)

Convert all fields with spectral coefficients to a regular Gaussian grid. The number of
latitudes of the resulting Gaussian grid is calculated from the triangular truncation by:
nlat = N IN T ((trunc 2 + 1.)/2.)
Use this operator to convert ERA40 data e.g. from TL159 to N80.

gp2sp

Gridpoint to spectral
Convert all Gaussian gridpoint fields to spectral coefficients. The triangular truncation
of the resulting spherical harmonics is calculated from the number of latitudes by:
trunc = (nlat 2 1)/ 3

gp2spl

Gridpoint to spectral (linear)

Convert all Gaussian gridpoint fields to spectral coefficients. The triangular truncation
of the resulting spherical harmonics is calculated from the number of latitudes by:
trunc = (nlat 2 1)/ 2
Use this operator to convert ERA40 data e.g. from N80 to TL159 instead of T106.

sp2sp

Spectral to spectral
Change the triangular truncation of all spectral fields. The operator performs downward
conversion by cutting the resolution. Upward conversions are achieved by filling in zeros.

spcut

Cut spectral wave number

Set the user defined wave numbers to zero.

Parameter

118

trunc

INTEGER

New spectral resolution

wnums

INTEGER

Comma separated list of wave numbers

Reference manual

Transformation

Example
To transform spectral coefficients from T106 to N80 Gaussian grid use:
cdo sp2gp i f i l e

ofile

To transform spectral coefficients from TL159 to N80 Gaussian grid use:

cdo s p 2 g p l i f i l e

ofile

2.11.2. WIND - Wind transformation

Synopsis
< operator > ifile ofile

Description
This module converts relative divergence and vorticity to U and V wind and vice versa.

Operators
dv2uv

Divergence and vorticity to U and V wind

Calculate U and V wind on a Gaussian grid from spherical harmonic coefficients of relative
divergence and vorticity. The divergence and vorticity need to have the names sd and
svo or code numbers 155 and 138. The number of latitudes of the resulting Gaussian
grid is calculated from the triangular truncation by:
nlat = N IN T ((trunc 3 + 1.)/2.)

dv2uvl

Divergence and vorticity to U and V wind (linear)

Calculate U and V wind on a Gaussian grid from spherical harmonic coefficients of relative
divergence and vorticity. The divergence and vorticity need tohave the names sd and svo
or code numbers 155 and 138. The number of latitudes of the resulting Gaussian grid is
calculated from the triangular truncation by:
nlat = N IN T ((trunc 2 + 1.)/2.)

uv2dv

U and V wind to divergence and vorticity

Calculate spherical harmonic coefficients of relative divergence and vorticity from U and
V wind. The U and V wind need to have the names u and v or the code numbers 131 and
132. The triangular truncation of the resulting spherical harmonics is calculated from
the number of latitudes by:
trunc = (nlat 2 1)/ 3

uv2dvl

U and V wind to divergence and vorticity (linear)

Calculate spherical harmonic coefficients of relative divergence and vorticity from U and
V wind. The U and V wind need to have the names u and v or the code numbers 131 and
132. The triangular truncation of the resulting spherical harmonics is calculated from
the number of latitudes by:
trunc = (nlat 2 1)/ 2

Example
Assume a dataset has at least spherical harmonic coefficients of divergence and vorticity. To transform
the spectral divergence and vorticity to U and V wind use:
cdo dv2uv i f i l e

ofile

119

Formatted I/O

Reference manual

2.12. Formatted I/O

This section contains modules to read and write ASCII data.
Here is a short overview of all operators in this section:
input
inputsrv
inputext

ASCII input
SERVICE ASCII input
EXTRA ASCII input

output
outputf
outputint
outputsrv
outputext

ASCII output
Formatted output
Integer output
SERVICE ASCII output
EXTRA ASCII output

120

Reference manual

Formatted I/O

2.12.1. INPUT - Formatted input

Synopsis
input,grid ofile
inputsrv ofile
inputext ofile

Description
This module reads time series of one 2D variable from standard input. All input fields need to have
the same horizontal grid. The format of the input depends on the chosen operator.

Operators
input

ASCII input
Reads fields with ASCII numbers from standard input and stores them in ofile. The
numbers read are exactly that ones which are written out by output.

inputsrv

SERVICE ASCII input

Reads fields with ASCII numbers from standard input and stores them in ofile. Each
field should have a header of 8 integers (SERVICE likely). The numbers that are read
are exactly that ones which are written out by outputsrv.

inputext

EXTRA ASCII input

Read fields with ASCII numbers from standard input and stores them in ofile. Each
field should have header of 4 integers (EXTRA likely). The numbers read are exactly
that ones which are written out by outputext.

Parameter
grid

STRING

Grid description file or name

Example
Assume an ASCII dataset contains a field on a global regular grid with 32 longitudes and 16 latitudes
(512 elements). To create a GRIB dataset from the ASCII dataset use:
cdo f grb input , r32x 16 o f i l e . grb < m y a s c i i d a t a

121

Formatted I/O

Reference manual

2.12.2. OUTPUT - Formatted output

Synopsis
output ifiles
outputf,format,nelem ifiles
outputint ifiles
outputsrv ifiles
outputext ifiles

Description
This module prints all values of all input datasets to standard output. All input fields need to have
the same horizontal grid. The format of the output depends on the chosen operator.

Operators
output

ASCII output
Prints all values to standard output. Each row has 6 elements with the C-style format
%13.6g.

outputf

Formatted output
Prints all values to standard output. The format and number of elements for each
row have to be specified by the parameters format and nelem.

outputint

Integer output
Prints all values rounded to the nearest integer to standard output.

outputsrv

SERVICE ASCII output

Prints all values to standard output. Each field with a header of 8 integers (SERVICE
likely).

outputext

EXTRA ASCII output

Prints all values to standard output. Each field with a header of 4 integers (EXTRA
likely).

Parameter
format

STRING

C-style format for one element (e.g. %13.6g)

nelem

INTEGER

Number of elements for each row

Example
To print all field elements of a dataset formatted with %8.4g and 8 values per line use:
cdo o u t p u t f , % 8 . 4 g , 8

ifile

Example result of a dataset with one field on 64 grid points:

261.7
250.6
273.4
267.5
275.3
282.3
292.9
293.8

122

262
252.6
266.2
267.4
275.5
284.5
290.5
292.6

257.8
253.9
259.8
272.2
276.4
280.3
293.9
291.2

252.5
254.8
261.6
266.7
278.4
280.3
292.6
292.6

248.8
252
257.2
259.6
282
280
292.7
293.2

247.7
246.6
253.4
255.2
269.6
281.5
292.8
292.8

246.3
249.7
251
272.9
278.7
284.7
294.1
291

246.1
257.9
263.7
277.1
279.5
283.6
293.6
291.2

Reference manual

Miscellaneous

2.13. Miscellaneous
This section contains miscellaneous modules which do not fit to the other sections before.
Here is a short overview of all operators in this section:
gridarea
gridweights

Grid cell area

Grid cell weights

gradsdes1
gradsdes2

GrADS data descriptor file (version 1 GRIB map)

GrADS data descriptor file (version 2 GRIB map)

smooth9

9 point smoothing

setrtoc
setrtoc2

Set range to constant

Set range to constant others to constant2

timsort

Sort over the time

const
random

Create a constant field

Create a field with random values

rotuvb

Backward rotation

mastrfu

Mass stream function

histcount
histsum
histmean
histfreq

Histogram
Histogram
Histogram
Histogram

wct

Windchill temperature

fdns

Frost days where no snow index per time period

strwin

Strong wind days index per time period

strbre

Strong breeze days index per time period

strgal

Strong gale days index per time period

hurr

Hurricane days index per time period

import amsr

Import AMSR binary files

import cmsaf

Import CM-SAF HDF5 files

import binary

Import binary data sets

count
sum
mean
frequency

123

Miscellaneous

Reference manual

2.13.1. GRIDCELL - Grid cell quantities

Synopsis
< operator > ifile ofile

Description
This module reads the grid cell area of the first grid from the input stream. If the grid cell area is
missing it will be computed from the grid description. Depending on the chosen operator the grid
cell area or weights are written to the output stream.

Operators
gridarea

Grid cell area

Writes the grid cell area to the output stream. If the grid cell area have to be
computed it is scaled with the earth radius to square meters.

gridweights

Grid cell weights

Writes the grid cell area weights to the output stream.

Environment
PLANET RADIUS

124

This variable is used to scale the computed grid cell areas to square meters. By
default PLANET RADIUS is set to an earth radius of 6371000 meter.

Reference manual

Miscellaneous

2.13.2. GRADSDES - GrADS data descriptor file

Synopsis
< operator > ifile

Description
Creates a GrADS data descriptor file. Supported file formats are GRIB, SERVICE, EXTRA and
IEG. For GRIB files the GrADS map file is also generated. For SERVICE and EXTRA files the grid
have to be specified with the CDO option -g <grid>. This module takes ifile in order to create
filenames for the descriptor (ifile.ctl) and the map (ifile.gmp) file. gradsdes is an alias for
gradsdes2.

Operators
gradsdes1

GrADS data descriptor file (version 1 GRIB map)

Creates a GrADS data descriptor file. Generated a machine specific version 1 GrADS
map file for GRIB datasets.

gradsdes2

GrADS data descriptor file (version 2 GRIB map)

Creates a GrADS data descriptor file. Generated a machine independent version 2
GrADS map file for GRIB datasets. This map file can be used only with GrADS
version 1.8 or newer.

Example
To create a GrADS data descriptor file from a GRIB dataset use:
cdo g r a d s d e s 2 i f i l e . grb

This will create a descriptor file with the name ifile.ctl and the map file ifile.gmp.
Assumed the input GRIB dataset has 3 variables over 12 time steps on a T21 grid. The contents of
the resulting GrADS data description file is approximately:
DSET i f i l e . grb
DTYPE GRIB
INDEX i f i l e . gmp
XDEF 64 LINEAR 0 . 0 0 0 0 0 0 5 . 6 2 5 0 0 0
YDEF 32 LEVELS 85.761 80.269 74.745 69.213 63.679 58.143
52.607 47.070 41.532 35.995 30.458 24.920
19.382 13.844 8.307 2.769
2.769
8.307
13.844 19.382 24.920 30.458 35.995 41.532
47.070 52.607 58.143 63.679 69.213 74.745
80.269 85.761
ZDEF 4 LEVELS 925 850 500 200
TDEF 12 LINEAR 1 2 : 0 0 Z1jan1987 1mo
TITLE i f i l e . grb T21 g r i d
OPTIONS y r e v
UNDEF 9e+33
VARS 3
geosp
0 129 ,1 ,0
s u r f a c e g e o p o t e n t i a l ( or og ra ph y )
[m2/ s 2 ]
t
4 130 ,99 ,0 temperature
[K]
tslm1
0 139 ,1 ,0
s u r f a c e temperature of land
[K]
ENDVARS

125

Miscellaneous

Reference manual

2.13.3. SMOOTH9 - 9 point smoothing

Synopsis
smooth9 ifile ofile

Description
Performs a 9 point smoothing on all fields with a regular quadrilateral grid. The result at each grid
point is a weighted average of the grid point plus the 8 surrounding points. The center point receives
a weight of 1.0, the points at each side and above and below receive a weight of 0.5, and corner points
receive a weight of 0.3. All 9 points are multiplied by their weights and summed, then divided by
the total weight to obtain the smoothed value. Any missing data points are not included in the sum;
points beyond the grid boundary are considered to be missing. Thus the final result may be the result
of an averaging with less than 9 points.

2.13.4. SETRANGE - Set range to constant

Synopsis
setrtoc,rmin,rmax,c ifile ofile
setrtoc2,rmin,rmax,c,c2 ifile ofile

Description
This module sets part of a field to a constant value or missing values. Which part of the field is set
depends on the chosen operator.

Operators
setrtoc

Set rangeto constant

c
if i(t, x) rmin i(t, x) rmax
o(t, x) =
i(t, x) if i(t, x) < rmin i(t, x) > rmax

setrtoc2

Set rangeto constant others to constant2

c if i(t, x) rmin i(t, x) rmax
o(t, x) =
c2 if i(t, x) < rmin i(t, x) > rmax

Parameter

126

rmin

FLOAT

Lower bound

rmax

FLOAT

Upper bound

FLOAT

New value - inside range

c2

FLOAT

New value - outside range

Reference manual

Miscellaneous

2.13.5. TIMSORT - Timsort

Synopsis
timsort ifile ofile

Description
Sorts the elements in ascending order over all time steps for every field position. After sorting it is:
(t1 < t2 ), x

o(t1 , x) <= o(t2 , x)

Example
To sort all field elements of a dataset over all time steps use:
cdo t i m s o r t i f i l e

ofile

2.13.6. VARGEN - Generate a field

Synopsis
const,const,grid ofile
random,grid ofile

Description
Generates a dataset with one field. The size of the field is specified by the user given grid description.
According to the chosen operator all field elements are constant or filled with random numbers.

Operators
const

Create a constant field

Creates a constant field. All field elements of the grid have the same value.

random

Create a field with random values

Creates a field with rectangularly distrubuted random numbers in the interval [0,1].

Parameter
const

FLOAT

Constant

grid

STRING

Target grid description file or name

127

Miscellaneous

Reference manual

2.13.7. ROTUV - Rotation

Synopsis
rotuvb,u,v,... ifile ofile

Description
This is a special operator for datsets with wind components on a rotated grid, e.g. data from the
regional model REMO. It performs a backward transformation of velocity components U and V from
a rotated spherical system to a geographical system.

Parameter
u,v,...

Pairs of zonal and meridional velocity components (use variable names or

code numbers)

STRING

Example
To transform the u and v velocity of a dataset from a rotated spherical system to a geographical
system use:
cdo rotuvb , u , v i f i l e

ofile

2.13.8. MASTRFU - Mass stream function

Synopsis
mastrfu ifile ofile

Description
This is a special operator for the post processing of the atmospheric general circulation model
ECHAM. It computes the mass stream function (code number 272). The input dataset have to
be a zonal mean of v-velocity (code number 132) on pressure levels.

Example
To compute the mass stream function from a zonal mean v-velocity dataset use:
cdo m a s t r f u i f i l e

128

ofile

Reference manual

Miscellaneous

2.13.9. HISTOGRAM - Histogram

Synopsis
< operator >,bounds ifile ofile

Description
This module creates bins for a histogram of the input data. The bins have to be adjacent and have
non-overlapping intervals. The user has to define the bounds of the bins. The first value is the lower
bound and the second value the upper bound of the first bin. The bounds of the second bin are
defined by the second and third value, aso. Only 2-dimensional input fields are allowed. The ouput
file contains one vertical level for each of the bins requested.

Operators
histcount

Histogram count
Number of elements in the bin range.

histsum

Histogram sum
Sum of elements in the bin range.

histmean

Histogram mean
Mean of elements in the bin range.

histfreq

Histogram frequency
Frequency of elements in the bin range.

Parameter
bounds

FLOAT

Comma separated list of the bin bounds (-inf and inf valid)

129

Miscellaneous

Reference manual

2.13.10. WCT - Windchill temperature

Synopsis
wct ifile1 ifile2 ofile

Description
Let ifile1 and ifile2 be time series of temperature and wind speed records, then a corresponding
time series of resulting windchill temperatures is written to ofile. The wind chill temperature calculation is only valid for a temperature of T <= 33 and a wind speed of v >= 1.39 m/s. Whenever
these conditions are not satisfied, a missing value is written to ofile. Note that temperature and
wind speed records have to be given in units of and m/s, respectively.

2.13.11. FDNS - Frost days where no snow index per time period
Synopsis
fdns ifile1 ifile2 ofile

Description
Let ifile1 be a time series of daily minimum temperatures TN and ifile2 be a corresponding series
of daily surface snow amounts. Then counted is the number of days where TN < 0 and the surface
snow amount is less than 1 cm. The temperature TN have to be given in units of Kelvin. The date
information of a time step in ofile is the date of the last contributing time step in ifile.

2.13.12. STRWIN - Strong wind days index per time period

Synopsis
strwin[,v] ifile ofile

Description
Let ifile be a time series of daily maximum horizontal wind speeds VX, then counted is the number
of days where VX > v. The horizontal wind speed v is an optional parameter with default v = 10.5
m/s. A further output variable is the maximum number of consecutive days with maximum wind
speed greater than or equal to v. Note that both VX and v have to be given in units of m/s. Also
note that the horizontal wind speed is defined as the square root of the sum of squares of the zonal
and meridional wind speeds. The date information of a time step in ofile is the date of the last
contributing time step in ifile.

Parameter
v

130

FLOAT

Horizontal wind speed threshold (m/s, default v = 10.5 m/s)

Reference manual

Miscellaneous

2.13.13. STRBRE - Strong breeze days index per time period

Synopsis
strbre ifile ofile

Description
Let ifile be a time series of daily maximum horizontal wind speeds VX, then counted is the number
of days where VX is greater than or equal to 10.5 m/s. A further output variable is the maximum
number of consecutive days with maximum wind speed greater than or equal to 10.5 m/s. Note that
VX is defined as the square root of the sum of squares of the zonal and meridional wind speeds and
have to be given in units of m/s. The date information of a time step in ofile is the date of the last
contributing time step in ifile.

2.13.14. STRGAL - Strong gale days index per time period

Synopsis
strgal ifile ofile

Description
Let ifile be a time series of daily maximum horizontal wind speeds VX, then counted is the number
of days where VX is greater than or equal to 20.5 m/s. A further output variable is the maximum
number of consecutive days with maximum wind speed greater than or equal to 20.5 m/s. Note that
VX is defined as the square root of the sum of square of the zonal and meridional wind speeds and
have to be given in units of m/s. The date information of a time step in ofile is the date of the last
contributing time step in ifile.

2.13.15. HURR - Hurricane days index per time period

Synopsis
hurr ifile ofile

Description
Let ifile be a time series of daily maximum horizontal wind speeds VX, then counted is the number
of days where VX is greater than or equal to 32.5 m/s. A further output variable is the maximum
number of consecutive days with maximum wind speed greater than or equal to 32.5 m/s. Note that
VX is defined as the square root of the sum of squares of the zonal and meridional wind speeds and
have to be given in units of m/s. The date information of a time step in ofile is the date of the last
contributing time step in ifile.

131

Miscellaneous

Reference manual

2.13.16. IMPORTAMSR - Import AMSR binary files

Synopsis
import amsr ifile ofile

Description
This operator imports gridded binary AMSR (Advanced Microwave Scanning Radiometer) data. The
binary data files are available from the AMSR ftp site (ftp://ftp.ssmi.com/amsre). Each file consists
of twelve (daily) or five (averaged) 0.25 x 0.25 degree grid (1440,720) byte maps. For daily files,
six daytime maps in the following order, Time (UTC), Sea Surface Temperature (SST), 10 meter
Surface Wind Speed (WSPD), Atmospheric Water Vapor (VAPOR), Cloud Liquid Water (CLOUD),
and Rain Rate (RAIN), are followed by six nighttime maps in the same order. Time-Averaged files
contain just the geophysical layers in the same order [SST, WSPD, VAPOR, CLOUD, RAIN]. More
information to the data is available on the AMSR homepage http://www.remss.com/amsr.

Example
To convert monthly binary AMSR files to netCDF use:
cdo f nc amsre yyyymmv5 amsre yyyymmv5 . nc

132

Reference manual

Miscellaneous

2.13.17. IMPORTCMSAF - Import CM-SAF HDF5 files

Synopsis
import cmsaf ifile ofile

Description
This operator imports gridded CM-SAF (Satellite Application Facility on Climate Monitoring) HDF5
files. CM-SAF exploits data from polar-orbiting and geostationary satellites in order to provide
climate monitoring products of the following parameters:
Cloud parameters: cloud fraction (CFC), cloud type (CTY), cloud phase (CPH), cloud top height,
pressure and temperature (CTH,CTP,CTT), cloud optical thickness (COT), cloud
water path (CWP).
Surface radiation components: Surface albedo (SAL); surface incoming (SIS) and net (SNS)
shortwave radiation; surface downward (SDL) and outgoing (SOL) longwave radiation, surface net longwave radiation (SNL) and surface radiation budget (SRB).
Top-of-atmosphere radiation components: Incoming (TIS) and reflected (TRS) solar radiative
flux at top-of-atmosphere. Emitted thermal radiative flux at top-of-atmosphere
(TET).
Water vapour: Vertically integrated water vapour (HTW), layered vertically integrated water
vapour and layer mean temperature and relative humidity for 5 layers (HLW), temperature and mixing ratio at 6 pressure levels.
Daily and monthly mean products can be ordered via the CM-SAF web page (www.cmsaf.eu). Products with higher spatial and temporal resolution, i.e. instantaneous swath-based products, are available on request ([email protected]). All products are distributed free-of-charge. More information on the data is available on the CM-SAF homepage (www.cmsaf.eu).
Daily and monthly mean products are provided in equal-area projections. CDO reads the projection
parameters from the metadata in the HDF5-headers in order to allow spatial operations like remapping. For spatial operations with instantaneous products on original satellite projection, additional
files with arrays of latitudes and longitudes are needed. These can be obtained from CM-SAF together
with the data.

Note
To use this operator, it is necessary to build CDO with HDF5 support (version 1.6 or higher). The
PROJ.4 library (version 4.6 or higher) is needed for full support of the remapping functionality.

Example
A typical sequence of commands with this operator could look like this:
cdo f nc remapbil , r3 60 x 18 0 i m p o r t c m s a f c m s a f p r o d u c t . hdf output . nc

(bilinear remapping to a predefined global grid with 1 deg resolution and conversion to netcdf).
If you work with CM-SAF data on original satellite project, an additional file with information on
geolocation is required, to perform such spatial operations:
cdo f nc remapbil , r7 20 x 36 0 s e t g r i d , c m s a f l a t l o n . h5 i m p o r t c m s a f cmsaf . hdf output . nc

Some CM-SAF data are stored as scaled integer values. For some operations, it could be desirable
(or necessary) to increase the accuracy of the converted products:
cdo b f 3 2 f nc fldmean s e l l o n l a t b o x , 0 , 1 0 , 0 , 1 0 remapbil , r7 20 x 36 0 \
i m p o r t c m s a f c m s a f p r o d u c t . hdf output . nc

133

Miscellaneous

Reference manual

2.13.18. IMPORTBINARY - Import binary data sets

Synopsis
import binary ifile ofile

Description
This operator imports gridded binary data sets via a GrADS data descriptor file. The GrADS data
descriptor file contains a complete description of the binary data as well as instructions on where to
find the data and how to read it. The descriptor file is an ASCII file that can be created easily with
a text editor. The general contents of a gridded data descriptor file are as follows:
Filename for the binary data
Missing or undefined data value
Mapping between grid coordinates and world coordinates
Description of variables in the binary data set

A detailed description of the components of a GrADS data descriptor file can be found in [GrADS].
Here is a list of the supported components: BYTESWAPPED, CHSUB, DSET, ENDVARS,
FILEHEADER, HEADERBYTES, OPTIONS, TDEF, TITLE, TRAILERBYTES, UNDEF, VARS,
XDEF, XYHEADER, YDEF, ZDEF

Example
To convert a binary data file to netCDF use:
cdo f nc i m p o r t b i n a r y i f i l e . c t l o f i l e . nc

Here is an example of a GrADS data descriptor file:

DSET i f i l e . b i n
OPTIONS s e q u e n t i a l
UNDEF 9e+33
XDEF 360 LINEAR 179.5 1
YDEF 180 LINEAR 89.5 1
ZDEF
1 LINEAR 1 1
TDEF
1 LINEAR 0 0 : 0 0 Z15jun1989 12 hr
VARS
1
param 1 99 d e s c r i p t i o n o f t h e v a r i a b l e
ENDVARS

The binary data file ifile.bin contains one parameter on a global 1 degree lon/lat grid written with
FORTRAN record length headers (sequential).

134

Reference manual

Climate indices

2.14. Climate indices

This section contains modules to compute the climate indices of daily temperature and precipitation
extremes.
Here is a short overview of all operators in this section:
eca cdd

Consecutive dry days index per time period

eca cfd

Consecutive frost days index per time period

eca csu

Consecutive summer days index per time period

eca cwd

Consecutive wet days index per time period

eca cwdi

Cold wave duration index wrt mean of reference period

eca cwfi

Cold-spell days index wrt 10th percentile of reference period

eca etr

Intra-period extreme temperature range

eca fd

Frost days index per time period

eca gsl

Growing season length index

eca hd

Heating degree days per time period

eca hwdi

Heat wave duration index wrt mean of reference period

eca hwfi

Warm spell days index wrt 90th percentile of reference period

eca id

Ice days index per time period

eca r10mm

Heavy precipitation days index per time period

eca r20mm

Very heavy precipitation days index per time period

eca r75p

Moderate wet days wrt 75th percentile of reference period

eca r75ptot

Precipitation percent due to R75p days

eca r90p

Wet days wrt 90th percentile of reference period

eca r90ptot

Precipitation percent due to R90p days

eca r95p

Very wet days wrt 95th percentile of reference period

eca r95ptot

Precipitation percent due to R95p days

eca r99p

Extremely wet days wrt 99th percentile of reference period

eca r99ptot

Precipitation percent due to R99p days

eca rr1

Wet days index per time period

eca rx1day

Highest one day precipitation amount per time period

eca rx5day

Highest five-day precipitation amount per time period

eca sdii

Simple daily intensity index per time period

135

Climate indices

Reference manual

eca su

Summer days index per time period

eca tg10p

Cold days percent wrt 10th percentile of reference period

eca tg90p

Warm days percent wrt 90th percentile of reference period

eca tn10p

Cold nights percent wrt 10th percentile of reference period

eca tn90p

Warm nights percent wrt 90th percentile of reference period

eca tr

Tropical nights index per time period

eca tx10p

Very cold days percent wrt 10th percentile of reference period

eca tx90p

Very warm days percent wrt 90th percentile of reference period

136

Reference manual

Climate indices

2.14.1. ECACDD - Consecutive dry days index per time period

Synopsis
eca cdd ifile ofile

Description
Let ifile be a time series of daily precipitation amounts RR, then counted is the largest number of
consecutive days where RR is less than 1 mm. A further output variable is the number of dry periods
of more than 5 days. The date information of a time step in ofile is the date of the last contributing
time step in ifile.

Example
To get the largest number of consecutive dry days of a time series of daily precipitation amounts use:
cdo e c a c d d r r f i l e

ofile

2.14.2. ECACFD - Consecutive frost days index per time period

Synopsis
eca cfd ifile ofile

Description
Let ifile be a time series of daily minimum temperatures TN, then counted is the largest number
of consecutive days where TN < 0 . Note that TN have to be given in units of Kelvin. The date
information of a time step in ofile is the date of the last contributing time step in ifile.

Example
To get the largest number of consecutive frost days of a time series of daily minimum temperatures
use:
cdo e c a c f d t n f i l e

ofile

137

Climate indices

Reference manual

2.14.3. ECACSU - Consecutive summer days index per time period

Synopsis
eca csu[,T] ifile ofile

Description
Let ifile be a time series of daily maximum temperatures TX, then counted is the largest number
of consecutive days where TX > T. The number T is an optional parameter with default T = 25 .
Note that TN have to be given in units of Kelvin, whereas T have to be given in degrees Celsius. The
date information of a time step in ofile is the date of the last contributing time step in ifile.

Parameter
T

FLOAT

Temperature threshold ( Celsius, default: T = 25 Celsius)

Example
To get the largest number of consecutive summer days of a time series of daily minimum temperatures
use:
cdo e c a c s u t x f i l e

ofile

2.14.4. ECACWD - Consecutive wet days index per time period

Synopsis
eca cwd ifile ofile

Description
Let ifile be a time series of daily precipitation amounts RR, then counted is the largest number of
consecutive days where RR is at least 1 mm. A further output variable is the number of wet periods
of more than 5 days. The date information of a time step in ofile is the date of the last contributing
time step in ifile.

Example
To get the largest number of consecutive wet days of a time series of daily precipitation amounts use:
cdo eca cwd r r f i l e

138

ofile

Reference manual

Climate indices

2.14.5. ECACWDI - Cold wave duration index wrt mean of reference period
Synopsis
eca cwdi[,nday[,T]] ifile1 ifile2 ofile

Description
Let ifile1 be a time series of daily minimum temperatures TN, and let ifile2 be the mean TNnorm
of daily minimum temperatures for any period used as reference. Then counted is the number of days
where, in intervals of at least nday consecutive days, TN < TNnorm - T. The numbers nday and T
are optional parameters with default nday = 6 and T = 5 . A further output variable is the number
of cold waves longer than or equal to nday days. Note that both TN and TNnorm have to be given
in the same units. The date information of a time step in ofile is the date of the last contributing
time step in ifile1.

Parameter
nday

INTEGER

Number of consecutive days (default: nday = 6)

FLOAT

Temperature offset ( Celsius, default: T = 5 Celsius)

Example
To compute the cold wave duration index of a time series of daily minimum temperatures use:
cdo e c a c w d i t n f i l e t n n o r m f i l e o f i l e

2.14.6. ECACWFI - Cold-spell days index wrt 10th percentile of reference period
Synopsis
eca cwfi[,nday] ifile1 ifile2 ofile

Description
Let ifile1 be a time series of daily mean temperatures TG, and ifile2 be the 10th percentile TGn10
of daily mean temperatures for any period used as reference. Then counted is the number of days
where, in intervals of at least nday consecutive days, TG < TGn10. The number nday is an optional
parameter with default nday = 6. A further output variable is the number of cold-spell periods longer
than or equal to nday days. Note that both TG and TGn10 have to be given in the same units. The
date information of a time step in ofile is the date of the last contributing time step in ifile1.

Parameter
nday

INTEGER

Number of consecutive days (default: nday = 6)

Example
To compute the number of cold-spell days of a time series of daily mean temperatures use:
cdo e c a c w f i t g f i l e t g n 1 0 f i l e o f i l e

139

Climate indices

Reference manual

2.14.7. ECAETR - Intra-period extreme temperature range

Synopsis
eca etr ifile1 ifile2 ofile

Description
Let ifile1 and ifile2 be time series of maximum and minimum temperatures TX and TN, respectively. Then the extreme temperature range is the difference of the maximum of TX and the minimum
of TN. Note that TX and TN have to be given in the same units. The date information of a time
step in ofile is the date of the last contributing time steps in ifile1 and ifile2.

Example
To get the intra-period extreme temperature range for two time series of maximum and minimum
temperatures use:
cdo e c a e t r t x f i l e

tnfile

ofile

2.14.8. ECAFD - Frost days index per time period

Synopsis
eca fd ifile ofile

Description
Let ifile be a time series of daily minimum temperatures TN, then counted is the number of days
where TN < 0 . Note that TN have to be given in units of Kelvin. The date information of a time
step in ofile is the date of the last contributing time step in ifile.

Example
To get the number of frost days of a time series of daily minimum temperatures use:
cdo e c a f d t n f i l e

140

ofile

Reference manual

Climate indices

2.14.9. ECAGSL - Growing season length index

Synopsis
eca gsl[,nday[,T[,fland]]] ifile1 ifile2 ofile

Description
Let ifile1 be a time series of daily mean temperatures TG, and ifile2 be a land-water mask. Then
counted are the number of days between the first occurrence of at least nday consecutive days with
TG > T and the first occurrence after 1 July of at least nday consecutive days with TG < T. The
numbers nday and T are optional parameter with default nday = 6 and T = 5 . The number fland
is an optional parameter with default value fland = 0.5 and denotes the fraction of a grid point that
have to be covered by land in order to be included in the calculation. A further output variable is the
start day of year of the growing season. Note that TG have to be given in units of Kelvin, whereas T
have to be given in degrees Celsius. The date information of a time step in ofile is the date of the
last contributing time step in ifile.

Parameter
nday

INTEGER

Number of consecutive days (default: nday = 6)

FLOAT

Temperature threshold (degree Celsius, default: T = 5 Celsius)

fland

FLOAT

Land fraction threshold (default: fland = 0.5)

Example
To get the growing season length of a time series of daily mean temperatures use:
cdo e c a g s l t g f i l e m a s k f i l e o f i l e

141

Climate indices

Reference manual

2.14.10. ECAHD - Heating degree days per time period

Synopsis
eca hd[,T1[,T2]] ifile ofile

Description
Let ifile be a time series of daily mean temperatures TG, then the heating degree days are defined
as the sum of T1 - TG, where only values TG < T2 are considered. If T1 and T2 are omitted, a
temperature of 17 is used for both parameters. If only T1 is given, T2 is set to T1. Note that TG
have to be given in units of kelvin, whereas T1 and T2 have to be given in degrees Celsius. The date
information of a time step in ofile is the date of the last contributing time step in ifile.

Parameter
T1

FLOAT

Temperature limit ( Celsius, default: T1 = 17 Celsius)

T2

FLOAT

Temperature limit ( Celsius, default: T2 = T1)

Example
To compute the heating degree days of a time series of daily mean temperatures use:
cdo e c a h d t g f i l e

ofile

2.14.11. ECAHWDI - Heat wave duration index wrt mean of reference period
Synopsis
eca hwdi[,nday[,T]] ifile1 ifile2 ofile

Description
Let ifile1 be a time series of daily maximum temperatures TX, and let ifile2 be the mean TXnorm
of daily maximum temperatures for any period used as reference. Then counted is the number of days
where, in intervals of at least nday consecutive days, TX > TXnorm + T. The numbers nday and T
are optional parameters with default nday = 6 and T = 5 . A further output variable is the number
of heat waves longer than or equal to nday days. Note that both TX and TXnorm have to be given
in the same units. The date information of a time step in ofile is the date of the last contributing
time step in ifile1.

Parameter
nday

INTEGER

Number of consecutive days (default: nday = 6)

FLOAT

Temperature offset ( Celsius, default: T = 5 Celsius)

Example
To compute the heat wave duration index of a time series of daily maximum temperatures use:
cdo e c a h w d i t x f i l e t x n o r m f i l e o f i l e

142

Reference manual

Climate indices

2.14.12. ECAHWFI - Warm spell days index wrt 90th percentile of reference period
Synopsis
eca hwfi[,nday] ifile1 ifile2 ofile

Description
Let ifile1 be a time series of daily mean temperatures TG, and ifile2 be the 90th percentile
TGn90 of daily mean temperatures for any period used as reference. Then counted is the number
of days where, in intervals of at least nday consecutive days, TG > TGn90. The number nday is an
optional parameter with default nday = 6. A further output variable is the number of warm-spell
periods longer than or equal to nday days. Note that both TG and TGn90 have to be given in the
same units. The date information of a time step in ofile is the date of the last contributing time
step in ifile1.

Parameter
nday

INTEGER

Number of consecutive days (default: nday = 6)

Example
To compute the number of warm-spell days of a time series of daily mean temperatures use:
cdo e c a h w f i t g f i l e t g n 9 0 f i l e o f i l e

2.14.13. ECAID - Ice days index per time period

Synopsis
eca id ifile ofile

Description
Let ifile be a time series of daily maximum temperatures TX, then counted is the number of days
where TX < 0 . Note that TX have to be given in units of Kelvin. The date information of a time
step in ofile is the date of the last contributing time step in ifile.

Example
To get the number of ice days of a time series of daily maximum temperatures use:
cdo e c a i d t x f i l e

ofile

143

Climate indices

Reference manual

2.14.14. ECAR10MM - Heavy precipitation days index per time period

Synopsis
eca r10mm ifile ofile

Description
Let ifile be a time series of daily precipitation amounts RR, then counted is the number of days
where RR is at least 10 mm. The date information of a time step in ofile is the date of the last
contributing time step in ifile.

Example
To get the number of days with precipitation greater than 10 mm for a time series of daily precipitation
amounts use:
cdo eca r10mm r r f i l e

ofile

2.14.15. ECAR20MM - Very heavy precipitation days index per time period
Synopsis
eca r20mm ifile ofile

Description
Let ifile be a time series of daily precipitation amounts RR, then counted is the number of days
where RR is at least 20 mm. The date information of a time step in ofile is the date of the last
contributing time step in ifile.

Example
To get the number of days with precipitation greater than 20 mm for a time series of daily precipitation
amounts use:
cdo eca r20mm r r f i l e

144

ofile

Reference manual

Climate indices

2.14.16. ECAR75P - Moderate wet days wrt 75th percentile of reference period
Synopsis
eca r75p ifile1 ifile2 ofile

Description
Let ifile1 be a time series of daily precipitation amounts RR, and ifile2 be the 75th percentile
RRn75 of daily precipitation amounts at wet days for any period used as reference. Then calculated
is the percentage of wet days with RR > RRn75. The date information of a time step in ofile is the
date of the last contributing time step in ifile1.

Example
To compute the percentage of wet days where the daily precipitation amount is greater than the 75th
percentile of the daily precipitation amount at wet days for a given reference period use:
cdo e c a r 7 5 p r r f i l e

rrn75file

ofile

2.14.17. ECAR75PTOT - Precipitation percent due to R75p days

Synopsis
eca r75ptot ifile1 ifile2 ofile

Description
Let ifile1 be a time series of daily precipitation amounts RR, and ifile2 be the 75th percentile
RRn75 of daily precipitation amounts at wet days for any period used as reference. Then calculated
is the ratio of the precipitation sum at wet days with RR > RRn75 to the total precipitation sum.
The date information of a time step in ofile is the date of the last contributing time step in ifile1.

145

Climate indices

Reference manual

2.14.18. ECAR90P - Wet days wrt 90th percentile of reference period

Synopsis
eca r90p ifile1 ifile2 ofile

Description
Let ifile1 be a time series of daily precipitation amounts RR, and ifile2 be the 90th percentile
RRn90 of daily precipitation amounts at wet days for any period used as reference. Then calculated
is the percentage of wet days with RR > RRn90. The date information of a time step in ofile is the
date of the last contributing time step in ifile1.

Example
To compute the percentage of wet days where the daily precipitation amount is greater than the 90th
percentile of the daily precipitation amount at wet days for a given reference period use:
cdo e c a r 9 0 p r r f i l e

rrn90file

ofile

2.14.19. ECAR90PTOT - Precipitation percent due to R90p days

Synopsis
eca r90ptot ifile1 ifile2 ofile

Description
Let ifile1 be a time series of daily precipitation amounts RR, and ifile2 be the 90th percentile
RRn90 of daily precipitation amounts at wet days for any period used as reference. Then calculated
is the ratio of the precipitation sum at wet days with RR > RRn90 to the total precipitation sum.
The date information of a time step in ofile is the date of the last contributing time step in ifile1.

146

Reference manual

Climate indices

2.14.20. ECAR95P - Very wet days wrt 95th percentile of reference period
Synopsis
eca r95p ifile1 ifile2 ofile

Description
Let ifile1 be a time series of daily precipitation amounts RR, and ifile2 be the 95th percentile
RRn95 of daily precipitation amounts at wet days for any period used as reference. Then calculated
is the percentage of wet days with RR > RRn95. The date information of a time step in ofile is the
date of the last contributing time step in ifile1.

Example
To compute the percentage of wet days where the daily precipitation amount is greater than the 95th
percentile of the daily precipitation amount at wet days for a given reference period use:
cdo e c a r 9 5 p r r f i l e

rrn95file

ofile

2.14.21. ECAR95PTOT - Precipitation percent due to R95p days

Synopsis
eca r95ptot ifile1 ifile2 ofile

Description
Let ifile1 be a time series of daily precipitation amounts RR, and ifile2 be the 95th percentile
RRn95 of daily precipitation amounts at wet days for any period used as reference. Then calculated
is the ratio of the precipitation sum at wet days with RR > RRn95 to the total precipitation sum.
The date information of a time step in ofile is the date of the last contributing time step in ifile1.

147

Climate indices

Reference manual

2.14.22. ECAR99P - Extremely wet days wrt 99th percentile of reference period
Synopsis
eca r99p ifile1 ifile2 ofile

Description
Let ifile1 be a time series of daily precipitation amounts RR, and ifile2 be the 99th percentile
RRn99 of daily precipitation amounts at wet days for any period used as reference. Then calculated
is the percentage of wet days with RR > RRn99. The date information of a time step in ofile is the
date of the last contributing time step in ifile1.

Example
To compute the percentage of wet days where the daily precipitation amount is greater than the 99th
percentile of the daily precipitation amount at wet days for a given reference period use:
cdo e c a r 9 9 p r r f i l e

rrn99file

ofile

2.14.23. ECAR99PTOT - Precipitation percent due to R99p days

Synopsis
eca r99ptot ifile1 ifile2 ofile

Description
Let ifile1 be a time series of daily precipitation amounts RR, and ifile2 be the 99th percentile
RRn99 of daily precipitation amounts at wet days for any period used as reference. Then calculated
is the ratio of the precipitation sum at wet days with RR > RRn99 to the total precipitation sum.
The date information of a time step in ofile is the date of the last contributing time step in ifile1.

148

Reference manual

Climate indices

2.14.24. ECARR1 - Wet days index per time period

Synopsis
eca rr1 ifile ofile

Description
Let ifile be a time series of daily precipitation amounts RR, then counted is the number of days
where RR is at least 1 mm. The date information of a time step in ofile is the date of the last
contributing time step in ifile.

Example
To get the number of wet days of a time series of daily precipitation amounts use:
cdo e c a r r 1 r r f i l e

ofile

2.14.25. ECARX1DAY - Highest one day precipitation amount per time period
Synopsis
eca rx1day[,mode] ifile ofile

Description
Let ifile be a time series of daily precipitation amounts RR, then the maximum of RR is written
to ofile. If the optional parameter mode is set to m, then maximum daily precipitation amounts
are determined for each month. The date information of a time step in ofile is the date of the last
contributing time step in ifile.

Parameter
mode

Operation mode (optional). If mode = m then maximum daily precipitation

amounts are determined for each month
STRING

Example
To get the maximum of a time series of daily precipitation amounts use:
cdo e c a r x 1 d a y r r f i l e

ofile

If you are interested in the maximum daily precipitation for each month, use:
cdo e c a r x 1 d a y ,m r r f i l e

ofile

Apart from metadata information, both operations yield the same as:
cdo timmax r r f i l e
cdo monmax r r f i l e

ofile
ofile

149

Climate indices

Reference manual

2.14.26. ECARX5DAY - Highest five-day precipitation amount per time period

Synopsis
eca rx5day[,x] ifile ofile

Description
Let ifile be a time series of 5-day precipitation totals RR, then the maximum of RR is written to
ofile. A further output variable is the number of 5 day period with precipitation totals greater than
x mm, where x is an optional parameter with default x = 50 mm. The date information of a time
step in ofile is the date of the last contributing time step in ifile.

Parameter
x

FLOAT

Precipitation threshold (mm, default: x = 50 mm)

Example
To get the maximum of a time series of 5-day precipitation totals use:
cdo e c a r x 5 d a y r r f i l e

ofile

Apart from metadata information, the above operation yields the same as:
cdo timmax r r f i l e

ofile

2.14.27. ECASDII - Simple daily intensity index per time period

Synopsis
eca sdii ifile ofile

Description
Let ifile be a time series of daily precipitation amounts RR, then the mean precipitation amount
at wet days (RR > 1 mm) is written to ofile. The date information of a time step in ofile is the
date of the last contributing time step in ifile.

Example
To get the daily intensity index of a time series of daily precipitation amounts use:
cdo e c a s d i i

rrfile

ofile

2.14.28. ECASU - Summer days index per time period

Synopsis
eca su[,T] ifile ofile

150

Reference manual

Climate indices

Description
Let ifile be a time series of daily maximum temperatures TX, then counted is the number of days
where TX > T. The number T is an optional parameter with default T = 25 . Note that TX have
to be given in units of Kelvin, whereas T have to be given in degrees Celsius. The date information
of a time step in ofile is the date of the last contributing time step in ifile.

Parameter
T

FLOAT

Temperature threshold (degree Celsius, default: T = 25 Celsius)

Example
To get the number of summer days of a time series of daily maximum temperatures use:
cdo e c a s u t x f i l e

ofile

151

Climate indices

Reference manual

2.14.29. ECATG10P - Cold days percent wrt 10th percentile of reference period
Synopsis
eca tg10p ifile1 ifile2 ofile

Description
Let ifile1 be a time series of daily mean temperatures TG, and ifile2 be the 10th percentile TGn10
of daily mean temperatures for any period used as reference. Then calculated is the percentage of
time where TG < TGn10. Note that both TG and TGn10 have to be given in the same units. The
date information of a time step in ofile is the date of the last contributing time step in ifile1.

Example
To compute the percentage of time where the daily mean temperature is less than the 10th percentile
of the daily mean temperature for a given reference period use:
cdo e c a t g 1 0 p t g f i l e t g n 1 0 f i l e o f i l e

2.14.30. ECATG90P - Warm days percent wrt 90th percentile of reference period
Synopsis
eca tg90p ifile1 ifile2 ofile

Description
Let ifile1 be a time series of daily mean temperatures TG, and ifile2 be the 90th percentile TGn90
of daily mean temperatures for any period used as reference. Then calculated is the percentage of
time where TG > TGn90. Note that both TG and TGn90 have to be given in the same units. The
date information of a time step in ofile is the date of the last contributing time step in ifile1.

Example
To compute the percentage of time where the daily mean temperature is greater than the 90th
percentile of the daily mean temperature for a given reference period use:
cdo e c a t g 9 0 p t g f i l e t g n 9 0 f i l e o f i l e

152

Reference manual

Climate indices

2.14.31. ECATN10P - Cold nights percent wrt 10th percentile of reference period
Synopsis
eca tn10p ifile1 ifile2 ofile

Description
Let ifile1 be a time series of daily minimum temperatures TN, and ifile2 be the 10th percentile
TNn10 of daily minimum temperatures for any period used as reference. Then calculated is the
percentage of time where TN < TNn10. Note that both TN and TNn10 have to be given in the same
units. The date information of a time step in ofile is the date of the last contributing time step in
ifile1.

Example
To compute the percentage of time where the daily minimum temperature is less than the 10th
percentile of the daily minimum temperature for a given reference period use:
cdo e c a t n 1 0 p t n f i l e t n n 1 0 f i l e o f i l e

2.14.32. ECATN90P - Warm nights percent wrt 90th percentile of reference period
Synopsis
eca tn90p ifile1 ifile2 ofile

Description
Let ifile1 be a time series of daily minimum temperatures TN, and ifile2 be the 90th percentile
TNn90 of daily minimum temperatures for any period used as reference. Then calculated is the
percentage of time where TN > TNn90. Note that both TN and TNn90 have to be given in the same
units. The date information of a time step in ofile is the date of the last contributing time step in
ifile1.

Example
To compute the percentage of time where the daily minimum temperature is greater than the 90th
percentile of the daily minimum temperature for a given reference period use:
cdo e c a t n 9 0 p t n f i l e t n n 9 0 f i l e o f i l e

153

Climate indices

Reference manual

2.14.33. ECATR - Tropical nights index per time period

Synopsis
eca tr[,T] ifile ofile

Description
Let ifile be a time series of daily minimum temperatures TN, then counted is the number of days
where TN > T. The number T is an optional parameter with default T = 20 . Note that TN have
to be given in units of Kelvin, whereas T have to be given in degrees Celsius. The date information
of a time step in ofile is the date of the last contributing time step in ifile.

Parameter
T

FLOAT

Temperature threshold ( Celsius, default: T = 20 Celsius)

Example
To get the number of tropical nights of a time series of daily minimum temperatures use:
cdo e c a t r t n f i l e

ofile

2.14.34. ECATX10P - Very cold days percent wrt 10th percentile of reference
period
Synopsis
eca tx10p ifile1 ifile2 ofile

Description
Let ifile1 be a time series of daily maximum temperatures TX, and ifile2 be the 10th percentile
TNx10 of daily maximum temperatures for any period used as reference. Then calculated is the
percentage of time where TX < TXn10. Note that both TX and TXn10 have to be given in the same
units. The date information of a time step in ofile is the date of the last contributing time step in
ifile1.

Example
To compute the percentage of time where the daily maximum temperature is less than the 10th
percentile of the daily maximum temperature for a given reference period use:
cdo e c a t x 1 0 p t x f i l e t x n 1 0 f i l e o f i l e

154

Reference manual

Climate indices

2.14.35. ECATX90P - Very warm days percent wrt 90th percentile of reference
period
Synopsis
eca tx90p ifile1 ifile2 ofile

Description
Let ifile1 be a time series of daily maximum temperatures TX, and ifile2 be the 90th percentile
TXn90 of daily maximum temperatures for any period used as reference. Then calculated is the
percentage of time where TX > TXn90. Note that both TX and TXn90 have to be given in the same
units. The date information of a time step in ofile is the date of the last contributing time step in
ifile1.

Example
To compute the percentage of time where the daily maximum temperature is greater than the 90th
percentile of the daily maximum temperature for a given reference period use:
cdo e c a t x 9 0 p t x f i l e t x n 9 0 f i l e o f i l e

155

Bibliography
[CDI]
Climate Data Interface, from the Max Planck Institute for Meteorologie
[CM-SAF]
Satellite Application Facility on Climate Monitoring, from the German Weather Service (Deutscher
Wetterdienst, DWD)
[ECHAM]
The atmospheric general circulation model ECHAM5, from the Max Planck Institute for Meteorologie
[GrADS]
Grid Analysis and Display System, from the Center for Ocean-Land-Atmosphere Studies (COLA)
[GRIB]
GRIB version 1, from the World Meteorological Organisation (WMO)
[HDF5]
HDF version 5, from the HDF Group
[INTERA]
INTERA Software Package, from the Max Planck Institute for Meteorologie
[MPIOM]
Ocean and sea ice model, from the Max Planck Institute for Meteorologie
[netCDF]
NetCDF Software Package, from the UNIDATA Program Center of the University Corporation for
Atmospheric Research
[PINGO]
The PINGO package, from the Model & Data group at the Max Planck Institute for Meteorologie
[REMO]
Regional Model, from the Max Planck Institute for Meteorologie
[PROJ.4]
Cartographic Projections Library, originally written by Gerald Evenden then of the USGS.
[SCRIP]
SCRIP Software Package, from the Los Alamos National Laboratory
[szip]
Szip compression software, developed at University of New Mexico.

156

A. Hints for PINGO user

Some CDO operators have the same name as in PINGO but the meaning is different. The following table
gives an overview of those operators.
Operator name

CDO

PINGO

min
max
daymean
daymin
daymax
monmean
monmin
monmax
seasmean

Minimum of two fields

Maximum of two fields
Daily mean
Daily minimum
Daily maximum
Monthly mean
Monthly minimum
Monthly maximum
Seasonally mean

Time minimum
Time maximum
Multi-year daily mean
Multi-year daily minimum
Multi-year daily maximum
Multi-year monthly mean
Multi-year monthly minimum
Multi-year monthly maximum
Multi-year seasonally mean

There are also some CDO operators with the same functionality as in PINGO but the name is different.
The following table gives an overview of those operators.

Maximum of two fields

Minimum of two fields
Field mean, min, max
Time mean, min, max
Daily mean, min, max
Monthly mean, min, max
Yearly mean, min, max
Running mean
Seasonally mean
Multi-year daily mean
Multi-year monthly mean
Multi-year seasonally mean

CDO

PINGO

max
min
fldmean, fldmin, fldmax
timmean, timmin, timmax
daymean, daymin, daymax
monmean, monmin, monmax
yearmean, yearmin, yearmax
runmean
seasmean
ydaymean
ymonmean
yseasmean

max2
min2
meanr minr, maxr
mean, min, max
daymeans, daymins, daymaxs
monmeans, monmins, monmaxs
yearmeans, yearmins, yearmaxs
runmeans
seasmeans
daymean
monmean
seasmean

157

B. Grid description examples

B.1. Example of a curvilinear grid description
Here is an example for the CDO description of a curvilinear grid. xvals/yvals describes the position of the
6x5 quadrilateral grid cells. The first 4 values of xbounds/ybounds are the corners of the first grid cell.
gridtype
gridsize
xsize
ysize
xvals

xbounds

yvals

ybounds

=
=
=
=
=

curvilinear
30
6
5
21 11
0
25
36 31
0
21
38
= 23 14 17
5
14
17
28 17 21
6
17
21
34 21 27
7
21
27
41 27 35
9
27
35
51 35 51
13
35
51
=
29
32
32
39
35
48
62
61
57
=
23
26
36
27
26
36
32
36
45
37
36
45
41
45
55
47
45
55
50
55
64
57
55
64
58
64
72
67
64
72

11
16
52
28
6
34
7
41
9
51
13
67
21
32
51
51
32
37
41
47
50
57
58
67
64
77

21
0
51

29
52
65

30
16
30
14
14
17
17
21
21
27
27
35
35
26
51
70
26
26
36
36
45
45
55
55
64
64

25
31
0
5
23
6
28
7
34
9
41
13
51
39
48
72
27
23
37
32
47
41
57
50
67
58

13
43
30
6
28
7
34
9
41
13
51
21
67
42
43
70
37
32
47
41
57
50
67
58
77
64

0
38
51
17
17
21
21
27
27
35
35
51
51
42
57
65
36
36
45
45
55
55
64
64
72
72

13
21
64
5
23
6
28
7
34
9
41
13
51

5
32
6
38
7
44
9
52
13
63

6
38
7
44
9
52
13
63
21
77

6
28
7
34
9
41
13
51
21
67

27
23
37
32
47
41
57
50
67
58

27
19
37
28
47
36
57
44
67
51

37
28
47
36
57
44
67
51
77
56

37
32
47
41
57
50
67
58
77
64

42
61
58

Figure B.1.: Orthographic and Robinson projection of the curvilinear grid

158

Grid description examples

Example description for unstructured grid cells

B.2. Example description for unstructured grid cells

Here is an example of the CDO description for unstructured grid cells. xvals/yvals describes the position
of 30 independent hexagonal grid cells. The first 6 values of xbounds/ybounds are the corners of the first
grid cell.
gridtype
gridsize
nvertex
xvals

xbounds

yvals

ybounds

= cell
= 30
= 6
= 36
144
0
= 339
0
20
72
92
144
164
216
236
268
345
20
92
164
236
=
58
32
58
=
41
11
19
11
19
11
19
11
19
19
41
19
19
19
19

36
126
72
0
16
36
88
108
160
180
232
252
288
324
15
87
159
231
58
0
58
53
19
11
19
11
19
11
19
11
7
53
41
41
41
41

0 18
18 108
72
162 108 144 162 126
36 144 108 144 180
0 288 288 309
21
0 339 344
36
16
0
0
93
72
51
56
108
88
72
72
165 144 123 128
180 160 144 144
237 216 195 200
252 232 216 216
288 272 252 252
324
36
36
15
36
57
52
36
108 129 124 108
180 201 196 180
252 273 268 252
32
0
0
58
32
0
58
32
0
0
32 58 32 58 32
71
71
53
41
41
53
41
19
7
19
11
7
41
53
41
19
7
19
11
7
41
53
41
19
7
19
11
7
41
53
41
19
7
19
11
7
11
19
7 11
71 71 53 41
53 41 19 11
53 41 19 11
53 41 19 11
53 41 19 11

54
90
72 90
72 108
21
51
340
0
93 123
52
72
165 195
124 144
237 267
196 216
288 304
308 324
36
36
108 108
180 180
252 252
308 303
0
0
32
0
58 32
41
41
19
7
41
41
19
7
41
41
19
7
41
41
19
7
11
19
19 11
53 71
53 71
53 71
53 71
19 41

180
54
36
72
0
144
72
216
144
288
216
309
324
108
180
252
324
324
58
0
32
53
11
53
11
53
11
53
11
41
7
71
71
71
71
53

72
344
144
56
216
128
288
200
288
304
108
180
252
324
345

0
324
72
36
144
108
216
180
267
288
87
159
231
303
340

0
324
72
36
144
108
216
180
272
288
57
129
201
273
324

71
19
71
19
71
19
71
19
53
19
53
53
53
53
41

71
7
71
7
71
7
71
7
41
11
41
41
41
41
19

53
11
53
11
53
11
53
11
19
7
41
41
41
41
11

Figure B.2.: Orthographic and Robinson projection of the unstructured grid cells

159

Operator index
A
abs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
acos . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
add . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
addc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
asin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
atan2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

58
58
60
59
58
60

C
cat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
chcode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
chlevel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
chlevelc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
chlevelv . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
chname . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
const . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
copy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
cos . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

D
dayavg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
daymax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
daymean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
daymin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
daypctl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
daystd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
daysum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
dayvar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
delcode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
delname . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
detrend . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
diff . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
diffv . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
div . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
divc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
divdpm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
divdpy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
dv2uv . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
dv2uvl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

E
eca
eca
eca
eca
eca
eca
eca
eca
eca

cdd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
cfd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
csu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
cwd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
cwdi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
cwfi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
etr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
fd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
gsl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

160

137
137
138
138
139
139
140
140
141

eca hd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
eca hwdi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
eca hwfi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
eca id . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
eca r10mm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
eca r20mm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
eca r75p . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
eca r75ptot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
eca r90p . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
eca r90ptot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
eca r95p . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
eca r95ptot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
eca r99p . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
eca r99ptot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
eca rr1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
eca rx1day . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
eca rx5day . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
eca sdii . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
eca su . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
eca tg10p . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
eca tg90p . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
eca tn10p . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
eca tn90p . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
eca tr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
eca tx10p . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
eca tx90p . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
enlarge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
ensavg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
ensmax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
ensmean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
ensmin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
enspctl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
ensstd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
enssum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
ensvar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
eq . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
eqc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
exp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
expr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
exprf . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

F
fdns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
fldavg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
fldmax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
fldmean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
fldmin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
fldpctl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
fldstd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
fldsum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

Operator index

Operator index

fldvar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

G
ge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
gec . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
genbic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
genbil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
gencon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
gencon2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
gendis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
genlaf . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
gennn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
gp2sp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
gp2spl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
gradsdes1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
gradsdes2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
gridarea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
griddes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
gridweights . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
gt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
gtc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

invertlev . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

40
41
109
109
109
109
109
109
109
118
118
125
125
124
23
124
40
41

H
histcount . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
histfreq . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
histmean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
histsum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
houravg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
hourmax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
hourmean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
hourmin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
hourpctl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
hourstd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
hoursum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
hourvar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
hurr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131

I
ifnotthen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
ifnotthenc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
ifthen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
ifthenc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
ifthenelse . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
import amsr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
import binary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
import cmsaf . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
info . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
infov . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
input . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
inputext . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
inputsrv . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
int . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
interpolate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
intlevel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
intntime . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
inttime . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
intyear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
invertlat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

le . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
lec . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ln . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
log10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
lt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ltc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

40
41
58
58
40
41

M
map . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
maskindexbox . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
masklonlatbox . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
maskregion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
mastrfu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
max . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
meravg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
merge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
mergetime . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
mermax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
mermean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
mermin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
merpctl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
merstd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
mersum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
mervar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
min . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
ml2hl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
ml2pl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
monadd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
monavg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
mondiv . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
monmax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
monmean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
monmin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
monmul . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
monpctl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
monstd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
monsub . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
monsum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
monvar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
mul . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
mulc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
muldpm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
muldpy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63

N
ndate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ne . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
nec . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
nint . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
nlevel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
nmon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
npar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ntime . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
nyear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

21
40
41
58
21
21
21
21
21

O
161

Operator index

Operator index

output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
outputext . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
outputf . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
outputint . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
outputsrv . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

122
122
122
122
122

P
pardes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
pow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

R
random . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
reci . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
regres . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
remap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
remapbic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
remapbil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
remapcon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
remapcon2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
remapdis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
remapeta . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
remaplaf . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
remapnn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
replace . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
rotuvb . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
runavg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
runmax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
runmean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
runmin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
runpctl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
runstd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
runsum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
runvar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79

S
seasavg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
seasmax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
seasmean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
seasmin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
seaspctl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
seasstd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
seassum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
seasvar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
selcode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
seldate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
selday . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
selgrid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
selgridname . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
selhour . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
selindexbox . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
sellevel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
sellevidx . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
sellonlatbox . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
selltype . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
selmon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
selname . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
selseas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
selsmon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

162

91
91
91
91
92
91
91
91
31
33
33
31
31
33
35
31
31
35
31
33
31
33
33

selstdname . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
seltabnum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
seltime . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
seltimestep . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
selyear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
selzaxis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
selzaxisname . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
setcalendar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
setcindexbox . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
setclonlatbox . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
setcode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
setctomiss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
setdate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
setday . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
setgatt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
setgatts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
setgrid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
setgridtype . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
setlevel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
setltype . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
setmisstoc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
setmissval . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
setmon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
setname . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
setpartab . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
setreftime . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
setrtoc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
setrtoc2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
setrtomiss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
settaxis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
settime . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
settunits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
setvrange . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
setyear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
setzaxis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
shifttime . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
showcode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
showdate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
showformat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
showlevel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
showltype . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
showmon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
showname . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
showstdname . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
showtime . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
showyear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
sin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
sinfo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
sinfop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
sinfov . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
smooth9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
sp2gp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
sp2gpl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
sp2sp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
spcut . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
splitcode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
splitday . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
splitgrid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

Operator index

Operator index

splithour . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
splitlevel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
splitmon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
splitname . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
splitseas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
splitsel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
splityear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
splitzaxis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
sqr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
sqrt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
strbre . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
strgal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
strwin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
sub . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
subc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
subtrend . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

T
tan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
timavg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
timmax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
timmean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
timmin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
timpctl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
timselavg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
timselmax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
timselmean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
timselmin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
timselpctl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
timselstd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
timselsum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
timselvar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
timsort . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
timstd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
timsum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
timvar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
trend . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

U
uv2dv . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
uv2dvl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

V
vct . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
vertavg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
vertmax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
vertmean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
vertmin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
vertstd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
vertsum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
vertvar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

23
76
76
76
76
76
76
76

W
wct . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130

Y
ydayavg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
ydaymax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
ydaymean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94

ydaymin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
ydaypctl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
ydaystd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
ydaysum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
ydayvar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
ydrunavg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
ydrunmax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
ydrunmean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
ydrunmin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
ydrunpctl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
ydrunstd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
ydrunsum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
ydrunvar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
yearavg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
yearmax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
yearmean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
yearmin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
yearpctl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
yearstd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
yearsum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
yearvar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
yhouravg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
yhourmax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
yhourmean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
yhourmin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
yhourstd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
yhoursum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
yhourvar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
ymonadd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
ymonavg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
ymondiv . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
ymonmax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
ymonmean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
ymonmin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
ymonmul . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
ymonpctl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
ymonstd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
ymonsub . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
ymonsum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
ymonvar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
yseasavg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
yseasmax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
yseasmean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
yseasmin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
yseaspctl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
yseasstd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
yseassum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
yseasvar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98

Z
zaxisdes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
zonavg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
zonmax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
zonmean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
zonmin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
zonpctl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
zonstd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
zonsum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

23
72
72
72
72
72
72
72

163

Operator index
zonvar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72

164

Operator index