SIDC/SILSO International sunspot number

1. The Daily Estimated Sunspot Number: provides the daily value for the the Estimated International Sunspot Number (EISN), which is obtained by a simple average over available sunspot counts from prompt stations in the SILSO network. The raw values from each station are scaled using their mean annual personal reduction coefficient, K, over the last elapsed year. Therefore, compared to the monthly international sunspot number (produced on the first day of each month), the accuracy of the EISN is lower because the calculation rests on a smaller number of stations and the K scaling coefficient is only an approximation of the true K coefficient of the month. The EISN is computed every few minutes and thus evolves continuously as new observations from our worldwide network are entered into our database. Therefore, the EISN gives a dynamical preview of the final sunspot number but is an ephemeral product. It should not be archived for long-term use. At the end of each month, those values are dropped and replaced by the provisional sunspot numbers from the full calculation for the corresponding month.

• Column 1 [1-4]: Gregorian Year
• Column 2 [6-8]: Gregorian Month
• Column 3 [9-10]: Gregorian Day
• Column 4 [12-19]: Decimal date
• Column 5 [21-23]: Estimated Sunspot Number
• Column 6 [25-29]: Estimated Standard Deviation
• Column 7 [31-33]: Number of Stations calculated
• Column 8 [35-37]: Number of Stations available

2. The Total Sunspot Number

2.1 The Daily Total Sunspot Number: ranges from 1/1/1818 until the last elapsed month (provisional values). It is derived by the formula: R= Ns + 10 * Ng, where Ns the number of spots and Ng the number of groups counted over the entire solar disk. No daily data are provided before 1818 because daily observations become too sparse in earlier years. Therefore, R. Wolf only compiled monthly means and yearly means for all years before 1818. In the TXT and CSV files, the missing values are marked by -1 (valid Sunspot Number are always positive).

The conventional 0.6 Zürich scale factor is not used anymore and A. Wolfer (Wolf's successor) is now defining the scale of the entire series. This puts the Sunspot Number at the scale of raw modern counts, instead of reducing it to the level of early counts by R. Wolf.

The error values correspond to the standard deviation of raw numbers provided by all stations. Before 1981, the errors are estimated with the help of an auto-regressive model based on the Poisson distribution of actual Sunspot Numbers. From 1981 onwards, the error value is the actual standard deviation of the sample of raw observations used to compute the daily value. The standard error of the daily sunspot number can be computed by: sigma/sqrt(N) where sigma is the listed standard deviation and N the number of observations for the day. Before 1981, the number of observations is set to 1, as the Sunspot Number was then essentially the raw Wolf number from the Zürich Observatory.

The data are available as text files in both ASCII and CSV (adapted for import e.g. in MS Excel) format, the delimiter in the CSV file is a semicolon ';'. The layout of the TXT and CSV files is the same, except for the final column, as described below. The character positions for the elements on each line in the TXT file is given in square brackets.

• Column 1 [1-4]: Gregorian Year
• Column 2 [6-7]: Gregorian Month
• Column 3 [9-10]: Gregorian Day
• Column 4 [12-19]: Date in fraction of year
• Column 5 [22-24]: Daily total sunspot number. A value of -1 indicates that no number is available for that day (missing value).
• Column 6 [26-30]: Daily standard deviation of the input sunspot numbers from individual stations.
• Column 7 [33-35]: Number of observations used to compute the daily value.
• Column 8 [37]: Definitive/provisional indicator. In the TXT file a blank indicates that the value is definitive and a '*' symbol indicates that the value is still provisional (subject to a possible revision, usually the last 3 to 6 months). In the CSV file a '1' indicates a definitive value and a '0' indicates a provisional value.

2.2 The Monthly Mean Total Sunspot Number: ranges from 1/1749 until the last elapsed month (provisional values). It is obtained by taking a simple arithmetic mean of the daily total sunspot number over all days of each calendar month. Monthly means are available only since 1749 because the original observations compiled by Rudolph Wolf were too sparse before that year (only yearly means are available back to 1700).

The error values are the monthly standard deviation of individual data and are derived from the daily values by: sigma(m)=sqrt(SUM(N(d)*sigma(d)^2)/SUM(N(d))) where sigma(d) is the standard deviation for a single day and N(d) is thenumber of observations for that day. The standard error on the monthly mean values can be computed by: sigma/sqrt(N) where sigma is the listed standard deviation and N the total number of observations in the month. NB: February 1824 does not contain any daily value. As it is the only month without data after 1749, the monthly mean value was interpolated by R. Wolf between the adjacent months.

The data are available as text files in both ASCII and CSV (adapted for import e.g. in MS Excel) format, the delimiter in the CSV file is a semicolon ';'. The layout of the TXT and CSV files is the same, except for the final column, as described below. The character positions for the elements on each line in the TXT file is given in square brackets.

• Column 1 [1-4]: Gregorian Year
• Column 2 [6-7]: Gregorian Month
• Column 3 [9-16]: Date in fraction of year for the middle of the corresponding month
• Column 4 [19-23]: Monthly mean total sunspot number.
• Column 5 [25-29]: Monthly mean standard deviation of the input sunspot numbers from individual stations.
• Column 6 [32-35]: Number of observations used to compute the monthly mean total sunspot number.
• Column 7 [36]: Definitive/provisional indicator. In the TXT file a blank indicates that the value is definitive and a '*' symbol indicates that the value is still provisional (subject to a possible revision, usually the last 3 to 6 months). In the CSV file a '1' indicates a definitive value and a '0' indicates a provisional value.

2.3 The Monthly Smoothed Total Sunspot Number ranges from 1/1749 until the last elapsed month (provisional values). It is derived by a "tapered-boxcar" running mean of monthly sunspot numbers over 13 months centred on the corresponding month (Smoothing function: equal weights = 1, except for first and last elements (-6 and +6 months) = 0.5, Normalization by 1/12 factor). There are no smoothed values for the first 6 months and last 6 months of the data series: columns 4, 5 and 6 are set to -1 (no data). This 13-month smoothed series is provided only for backward compatibility with a large number of past publications and methods resting on this smoothed series. It has thus become a base standard (e.g. for the conventional definition of the times of minima and maxima of solar cycles). However, a wide range of other smoothing functions can be used, often with better low-pass filtering and anti-aliasing properties. As the optimal filter choice depends on the application, we thus invite users to start from the monthly mean Sunspot Numbers and apply the smoothing function that is most appropriate for their analyses. The classical smoothed series included here should only be used when direct comparisons with past published analyses must be made.

The standard deviations in this files are obtained from the weighted mean of the variances of the 13 months in the running mean value: sigma(ms)=sqrt(SUM(weight(M)*sigma(M)^2)/SUM(weight(M)) where sigma(M) is the standard deviation for a single month, weight(M) is 1 or 0.5 and M=13 in this case. As successive monthly means are highly correlated, the standard error on the smoothed values can be estimated by the same formula as for a single month: sigma/sqrt(N) where sigma is the listed standard deviation and N the total number of observations in the month. The number of observations given in column 6 is the number of observations of the corresponding (middle) month: same value SUM N(d) as in the monthly mean file. This thus gives a smoothed mean of monthly standard deviations, i.e. with the same low-pass filtering as the data value itself. Further autocorrelation analyses will be needed to derive a conversion of this standard deviation to a standard error of the 13-month smoothed number.

The data are available as text files in both ASCII and CSV (adapted for import e.g. in MS Excel) format, the delimiter in the CSV file is a semicolon ';'. The layout of the TXT and CSV files is the same, except for the final column, as described below. The character positions for the elements on each line in the TXT file is given in square brackets.

• Column 1 [1-4]: Gregorian Year
• Column 2 [6-7]: Gregorian Month
• Column 3 [9-16]: Date in fraction of year for the middle of the corresponding month
• Column 4 [19-23]: Monthly smoothed total sunspot number.
• Column 5 [25-29]: Monthly mean standard deviation of the input sunspot numbers.
• Column 6 [32-35]: Number of observations used to compute the corresponding monthly mean total sunspot number.
• Column 7 [37]: Definitive/provisional indicator. In the TXT file a blank indicates that the value is definitive and a '*' symbol indicates that the value is still provisional (subject to a possible revision, usually the last 3 to 6 months). In the CSV file a '1' indicates a definitive value and a '0' indicates a provisional value.

2.4 The Yearly Mean Total Sunspot Number ranges from 1700 until the last elapsed year. It is obtained by taking a simple arithmetic mean of the daily total sunspot number over all days of each year. (NB: in early years in particular before 1749, the means are computed on only a fraction of the days in each year because on many days, no observation is available).

The yearly standard deviation of individual data is derived from the daily values by the same formula as the monthly means: sigma(m)=sqrt(SUM(N(d)*sigma(d)^2)/SUM(N(d))) where sigma(d) is the standard deviation for a single day and N(d) is the number of observations for that day. The standard error on the yearly mean values can be computed by: sigma/sqrt(N) where sigma is the listed standard deviation and N the total number of observations in the year. NB: this standard error gives a measure of the precision, i.e. the sensitivity of the yearly value to different samples of daily values with random errors. The uncertainty on the mean (absolute accuracy) is only determined on longer time scales, and is thus not given here for individual yearly values.

The data are available as text files in both ASCII and CSV (adapted for import e.g. in MS Excel) format, the delimiter in the CSV file is a semicolon ';'. The layout of the TXT and CSV files is the same, except for the final column, as described below. The character positions for the elements on each line in the TXT file is given in square brackets.

• Column 1 [1-6]: Gregorian calendar year (mid-year date)
• Column 2 [9-13]: Yearly mean total sunspot number.
• Column 3 [15-19]: Yearly mean standard deviation of the input sunspot numbers from individual stations.
• Column 4 [22-26]: Number of observations used to compute the yearly mean total sunspot number.
• Column 5 [28]: Definitive/provisional indicator. In the TXT file a blank indicates that the value is definitive and a '*' symbol indicates that the value is still provisional (subject to a possible revision, usually the last 3 to 6 months). In the CSV file a '1' indicates a definitive value and a '0' indicates a provisional value.

3. The Hemispheric Sunspot Numbers

3.1 The Daily Total and Hemispheric Sunspot Numbers range from 1/1/1992 until the last elapsed month (provisional values). They are derived by the formula: R= Ns + 10 * Ng, with Ns the number of spots and Ng the number of groups counted either over the entire solar disc (total), the North hemisphere or South hemisphere (based on the sunspot group heliographic latitude). The North and South numbers are always normalized to the total number, which is the global scaling reference. The production of the hemispheric numbers together with the international total Sunspot Number started only in 1992.

The conventional 0.6 Zürich scale factor is not used anymore and A. Wolfer (Wolf's successor) is now defining the scale of the entire series. This puts the Sunspot Number at the scale of raw modern counts, instead of reducing it to the level of early counts by R. Wolf.

The error values correspond to the standard deviation of raw numbers provided by all stations. The error value for the total number is the actual standard deviation of the sample of raw observations used to compute the daily value. As the actual standard deviations of raw daily hemispheric counts were not archived in the past, we derive an estimate of these standard deviations with the help of the same auto-regressive model based on the Poisson distribution of actual Sunspot Numbers.

The data are available as text files in both ASCII and CSV (adapted for import e.g. in MS Excel) format, the delimiter in the CSV file is a semicolon ';'. The layout of the TXT and CSV files is the same, except for the final column, as described below. The character positions for the elements on each line in the TXT file is given in square brackets.

• Column 1 [1-4]: Gregorian Year
• Column 2 [6-7]: Gregorian Month
• Column 3 [9-10]: Gregorian Day
• Column 4 [12-19]: Date in fraction of year
• Column 5 [22-24]: Daily total sunspot number.
• Column 6 [26-28]: Daily North sunspot number.
• Column 7 [30-32]: Daily South sunspot number.
• Column 8 [35-39]: Standard deviation of raw daily total sunspot data
• Column 9 [41-45]: Standard deviation of raw daily North sunspot data
• Column 10 [46-51]: Standard deviation of raw daily South sunspot data
• Column 11 [54-56]: Number of observations in daily total sunspot number
• Column 12 [58-60]: Number of observations in daily North sunspot number (not determined yet: -1)
• Column 13 [62-64]: Number of observations in daily South sunspot number (not determined yet: -1)
• Column 14 [66]: Definitive/provisional indicator. In the TXT file a blank indicates that the value is definitive and a '*' symbol indicates that the value is still provisional (subject to a possible revision, usually the last 3 to 6 months). In the CSV file a '1' indicates a definitive value and a '0' indicates a provisional value.

3.2 The Monthly Mean Hemispheric Sunspot Numbers range from 1/1992 to the last elapsed month (provisional values). Monthly mean North/South sunspot numbers obtained by taking a simple arithmetic mean of the daily North/South sunspot numbers over all days of each calendar month. Each sunspot group is counted in one of the hemispheres based on its heliographic latitude. NB: Hemispheric numbers are always normalized to the corresponding total monthly mean sunspot number (cf. file monthssn.dat)

For total numbers, the monthly standard deviation of individual data is derived from the daily values by: sigma(m)=sqrt(SUM(N(d)*sigma(d)^2)/SUM(N(d))) where sigma(d) is the standard deviation for a single day and N(d) is the number of observations for that day. For hemispheric values, the errors are derived from daily values by: sigma(m)=sqrt(SUM(sigma(d)^2)/Nobs)

The data are available as text files in both ASCII and CSV (adapted for import e.g. in MS Excel) format, the delimiter in the CSV file is a semicolon ';'. The layout of the TXT and CSV files is the same, except for the final column, as described below. The character positions for the elements on each line in the TXT file is given in square brackets.

• Column 1 [1-4]: Gregorian Year
• Column 2 [6-7]: Gregorian Month
• Column 3 [9-16]: Date in fraction of year for the middle of the corresponding month
• Column 4 [19-23]: Monthly mean total sunspot number.
• Column 5 [25-29]: North monthly mean sunspot number.
• Column 6 [31-35]: South monthly mean sunspot number.
• Column 7 [38-42]: Monthly mean standard deviation of total sunspot number data
• Column 8 [44-48]: Monthly mean standard deviation of North sunspot number data
• Column 9 [50-54]: Monthly mean standard deviation of South sunspot number data
• Column 10 [57-60]: Number of observations used to compute the monthly mean total sunspot number.
• Column 11 [62-65]: Number of observations for the monthly mean North sunspot number (not used yet)
• Column 12 [67-70]: Number of observations for the monthly mean South sunspot number (not used yet)
• Column 13 [72]: Definitive/provisional indicator. In the TXT file a blank indicates that the value is definitive and a '*' symbol indicates that the value is still provisional (subject to a possible revision, usually the last 3 to 6 months). In the CSV file a '1' indicates a definitive value and a '0' indicates a provisional value.

3.3 The Monthly Smooth Hemispheric Sunspot Numbers ranges from 1/1992 to the last elapsed month (provisional values). The 13-month smoothed monthly sunspot numbers are derived by a "tapered-boxcar" running mean of monthly hemispheric sunspot numbers over 13 months centred on the corresponding month (Smoothing function: equal weights = 1, except for first and last elements (-6 and +6 months) = 0.5, Normalization by 1/12 factor). This is the same smoothing as the standard smoothing applied to the total monthly sunspot numbers. There are no smoothed values for the first 6 months and last 6 months of the file: columns 4, 5 and 6 are set to -1 (no data).

This 13-month smoothed series is provided only for backward compatibility with a large number of past publications and methods resting on this smoothed series. It has thus become a base standard (e.g. for the conventional definition of the times of minima and maxima of solar cycles). However, a wide range of other smoothing functions can be used, often with better low-pass filtering and anti-aliasing properties. As the optimal filter choice depends on the application, we thus invite users to start from the monthly mean Sunspot Numbers and apply the smoothing function that is most appropriate for their analyses. The classical smoothed series included here should only be used when direct comparisons with past published analyses must be made.

For total Sunspot Numbers, the standard deviations in this files are obtained from the weighted mean of the variances of the 13 months in the running mean value: sigma(ms)=sqrt(SUM(weight(M)*sigma(M)^2)/SUM(weight(M)) where sigma(M) is the standard deviation for a single month, weight(M) is 1 or 0.5 and M is over 13 months in this case. The standard deviations for the hemispheric values are estimated by the weighted mean (squared) of the error on the monthly mean total sunspot number. As successive monthly means are highly correlated, the standard error on the smoothed values can be estimated by the same formula as for a single month: sigma/sqrt(N) where sigma is the listed standard deviation and N the total number of observations in the month. The number of observations given in column 6 is the number of observations of the corresponding (middle) month: same value SUM N(d) as in the monthly mean file. This gives a smoothed mean of monthly standard deviations, i.e. with the same low-pass filtering as the data value itself. Further autocorrelation analyses will be needed to derive a conversion of this standard deviation to a standard error of the 13-month smoothed number.

The data are available as text files in both ASCII and CSV (adapted for import e.g. in MS Excel) format, the delimiter in the CSV file is a semicolon ';'. The layout of the TXT and CSV files is the same, except for the final column, as described below. The character positions for the elements on each line in the TXT file is given in square brackets.

• Column 1 [1-4]: Gregorian Year
• Column 2 [6-7]: Gregorian Month
• Column 3 [9-16]: Date in fraction of year for the middle of the corresponding month
• Column 4 [19-23]: Monthly smoothed total sunspot number.
• Column 5 [25-29]: Monthly smoothed North sunspot number.
• Column 6 [31-35]: Monthly smoothed South sunspot number.
• Column 7 [38-42]: Monthly mean standard deviation of the raw total sunspot number data.
• Column 8 [44-48]: Monthly mean standard deviation of the raw North sunspot number data.
• Column 9 [50-54]: Monthly mean standard deviation of the raw South sunspot number data.
• Column 10 [57-60]: Number of observations used to compute the corresponding monthly mean total sunspot number.
• Column 11 [62-65]: Number of observations for the monthly mean north sunspot number (not used yet)
• Column 12 [67-70]: Number of observations for the monthly mean south sunspot number (not used yet)
• Column 13 [72]: Definitive/provisional indicator. In the TXT file a blank indicates that the value is definitive and a '*' symbol indicates that the value is still provisional (subject to a possible revision, usually the last 3 to 6 months). In the CSV file a '1' indicates a definitive value and a '0' indicates a provisional value.