Station Datasets
Station data files organise data in blocks with each block representing a field, with the rows representing time, and the columns stations. The tags are similar to those for gridded data . Compulsory tags are:
- cpt:nrow= number of time steps
- cpt:ncol= number of stations
- cpt:row=T (representing time)
- cpt:col=station
The best way to organise a station dataset is to have successive months in each row. The number of time steps will then be equal to the 12 x the number of years of data available (plus the number of any additional months available). With all months present in the file, CPT will attempt to detect the season automatically, or prompt for the season. Organised in this way, the same input file can then be used for any season.
If the first twelve time steps are not consecutive months, CPT will expect the number of time steps to be equal to the number of years available multiplied by the number of lags. For example, if data are available for January and February 1971 to 2000 there are two months (one lagged field) and 30 years of data, making a total of 60 time steps. Alternatively, lagged fields can be included as additional fields, and the number of time steps indicated as 30 for each. If there are multiple fields then only the number of time steps for the current field should be listed. If any of the months are missing in the file, cpt:nrow should specify the number of time steps that are available. For example, if the data for February 1980 are missing then cpt:nrow should indicate 59 time steps. Alternatively, if the month is included in the file but all the stations have missing values listed then cpt:nrow should indicate that 60 time steps are available. See the section on missing values for further details.
Optional tags are:
- cpt:field= abbreviated name of the field variable
- cpt:units= units in which the data are stored
- cpt:missing= missing value flag
Any additional tags are ignored. For further details, see the section on CPT tags .
Immediately beneath the tag-line should be a line containing the names, or some kind of unique identifying reference, for each of the ncol stations. The name for each station must not be longer than 16 characters, and should contain no spaces. The latitudes and longitudes of each station then follow on the next two lines. The line for the longitudes must begin with "cpt:X" and that for the latitudes with "cpt:Y". It does not matter whether the latitudes or the longitudes are listed first. As with the gridded data , latitudes must be between 90 and -90 with negative latitudes representing the southern hemisphere, and longitudes must be between -180 and 360, with negative longitudes representing the western hemisphere. Additional tagged lines may follow, as in the example below, which includes a line for elevation and one for province. These lines are currently ignored, and are not included in any output files.
The data are arranged in columns by station as the "cpt:col" tag indicates, but the first column must contain the corresponding date. The date format is the same as for gridded data . The seasonal average format with the "/" separator is permissible. If the data are input as consecutive months, the file should look something like the following:
xmlns:cpt=http://iri.columbia.edu/CPT/v10/
cpt:field=prcp, cpt:nrow=14, cpt:ncol=6, cpt:row=T, cpt:col=station, cpt:units=mm/month, cpt:missing=-2.
A B C D E F
cpt:X -87.05 -87.88 -88.13 -87.58 -86.32 -87.62
cpt:Y 31.07 30.55 32.83 32.70 31.95 34.75
cpt:elev 85 m 23 m 125 m 220 m 594 m 536 m
1981-01 1.46 0.94 1.42 1.94 0.92 1.45
1981-02 11.20 11.59 8.02 6.01 7.29 4.03
1981-03 4.61 2.59 6.94 7.97 5.58 3.70
1981-04 0.50 1.24 6.93 5.89 2.77 2.38
1981-05 5.25 11.88 2.49 3.45 3.30 3.65
1981-06 3.92 2.51 2.95 4.08 3.09 5.35
1981-07 7.96 5.05 5.00 2.43 4.26 6.07
1981-08 5.29 6.28 6.80 6.01 6.65 6.57
1981-09 1.84 1.11 2.60 0.74 1.69 1.33
1981-10 1.63 1.90 3.52 5.20 2.60 3.11
1981-11 2.02 0.81 0.80 0.75 2.08 3.98
1981-12 7.52 5.88 5.99 6.92 7.86 1.85
1982-01 6.42 3.23 6.50 7.83 4.72 7.46
1982-02 12.19 9.58 8.80 7.41 9.15 4.08
For annually sequenced data:
xmlns:cpt=http://iri.columbia.edu/CPT/v10/
cpt:field=prcp, cpt:nrow=8, cpt:ncol=6, cpt:row=T, cpt:col=station, cpt:units=mm/month, cpt:missing=-2.
A B C D E F
cpt:X -87.05 -87.88 -88.13 -87.58 -86.32 -87.62
cpt:Y 31.07 30.55 32.83 32.70 31.95 34.75
cpt:elev 85 m 23 m 125 m 220 m 594 m 536 m
1981-01 1.46 0.94 1.42 1.94 0.92 1.45
1982-01 6.42 3.23 6.50 7.83 4.72 7.46
1983-01 4.75 6.60 3.82 4.23 3.46 2.57
1984-01 4.49 5.74 4.20 3.37 -2.00 2.64
1985-01 5.81 5.08 3.32 3.96 4.48 3.14
1986-01 6.76 3.80 1.15 1.31 3.42 0.60
1987-01 6.70 6.35 7.04 9.08 4.38 3.52
1988-01 7.20 4.38 3.56 6.86 -2.00 2.51
With an additional lagged field, the file may look like this:
xmlns:cpt=http://iri.columbia.edu/CPT/v10/
cpt:field=prcp, cpt:nrow=8, cpt:ncol=6, cpt:row=T, cpt:col=station, cpt:units=mm/month, cpt:missing=-2.
A B C D E F
cpt:X -87.05 -87.88 -88.13 -87.58 -86.32 -87.62
cpt:Y 31.07 30.55 32.83 32.70 31.95 34.75
cpt:elev 85 m 23 m 125 m 220 m 594 m 536 m
1981-01 1.46 0.94 1.42 1.94 0.92 1.45
1981-02 11.20 11.59 8.02 6.01 7.29 4.03
1982-01 6.42 3.23 6.50 7.83 4.72 7.46
1982-02 12.19 9.58 8.80 7.41 9.15 4.08
1983-01 4.75 6.60 3.82 4.23 3.46 2.57
1983-02 8.70 11.61 6.61 12.49 8.39 5.20
1984-01 4.49 5.74 4.20 3.37 -2.00 2.64
1984-02 2.65 5.95 4.27 4.33 4.41 3.63
It is possible for some of the dates to be missing, although there are some restrictions. For further details see the section on missing values .
Any additional fields are included immediately after the data, starting with a new tag-line. All fields must have the same number of lagged fields, but the lags do not have to be the same. The number of stations, their locations, and the missing value flag can all differ from the first field. Even the dates can differ, but the sequencing must be identical, and the first time steps of each fild are assumed to correspond. An example is shown below.
xmlns:cpt=http://iri.columbia.edu/CPT/v10/
cpt:nfields=2
cpt:field=prcp, cpt:nrow=8, cpt:ncol=5, cpt:row=T, cpt:col=station, cpt:units=mm/month, cpt:missing=-2.
A B C D E
cpt:X 122.006 121.050 121.633 120.600 121.557
cpt:Y 14.102 14.083 18.367 16.417 15.757
cpt:elev 5 m 10 m 1503 m 25 m 6 m
cpt:Province Z Y X W V
2000-01 -2.0 121.5 124.0 98.0 95.5
2000-02 345.5 87.5 224.0 173.5 34.5
2001-01 428.5 93.5 301.0 104.0 -2.0
2001-02 341.0 126.0 156.0 130.0 62.5
2002-01 -2.0 34.0 185.5 75.5 71.5
2002-02 117.5 72.0 94.0 216.0 67.5
2003-01 143.5 34.5 241.0 134.5 4.5
2003-02 355.0 81.5 137.5 62.5 7.0
cpt:field=temp, cpt:nrow=10, cpt:ncol=4, cpt:row=T, cpt:col=station, cpt:units=C, cpt:missing=-2.
A B C D
cpt:X 122.006 121.050 121.633 120.600
cpt:Y 14.102 14.083 18.367 16.417
cpt:elev 5 m 10 m 1503 m 25 m
cpt:Province Z Y X W
2000-01 -2.0 23.3 14.4 22.1
2000-02 20.4 24.9 17.6 23.6
2001-01 23.3 24.8 15.8 25.4
2001-02 25.7 25.7 13.0 24.2
2002-01 -2.0 24.6 17.2 25.7
2002-02 25.8 22.7 18.3 23.8
2003-01 27.4 26.3 19.6 25.5
2003-02 24.0 24.1 16.4 23.6
2004-01 26.6 25.2 16.9 24.3
2004-02 24.4 23.7 16.3 22.1
In the example above the second field has more years than the first. CPT will recognise that there are at most five years in the file and will treat 2004 in the first field as missing data.
To verify station forecasts in CPT using the Probabilistic Forecast Verification (PFV) option the X file needs to contain the tag cpt:ncat=3, indicating that the file contains probabilities for three categories. The first block of data should then contain the probabilities for category 1, which is the below-normal category. The category number should be indicated using the cpt:C tag, and, optionally, the climatological probability for the category, by using the cpt:clim_prob tag. The forecasts for categories 2 and 3 should then follow immediately, as if they were unstacked fields. For example, a station probabilistic forecast input file might look something like the following:
xmlns:cpt=http://iri.columbia.edu/CPT/v10/
cpt:ncats=3
cpt:field=prcp, cpt:C=1, cpt:clim_prob=0.333333333333, cpt:nrow=3, cpt:ncol=4, cpt:row=T, cpt:col=station, cpt:units=%, cpt:missing=-9999
A B C D
cpt:Y 18.2 17.1 13.2 15.2
cpt:X -63.1 -61.8 -59.6 -55.2
2000-01/03 -9999 -9999 -9999 -9999
2001-01/03 50 50 45 40
2002-01/03 35 25 25 15
cpt:C=2, cpt:clim_prob=0.333333333334
A B C D
cpt:Y 18.2 17.1 13.2 15.2
cpt:X -63.1 -61.8 -59.6 -55.2
2000-01/03 -9999 -9999 -9999 -9999
2001-01/03 35 40 45 35
2002-01/03 40 35 30 25
cpt:C=3, cpt:clim_prob=0.333333333333
A B C D
cpt:Y 18.2 17.1 13.2 15.2
cpt:X -63.1 -61.8 -59.6 -55.2
2000-01/03 -9999 -9999 -9999 -9999
2001-01/03 25 25 20 40
2002-01/03 25 40 45 60