Provisional look at solar constant 1923 to 1954
1. A single data point is present 1957 which is probably a dataset error in original transcription.
I decided to show a part of a larger work I am intending to post which is about the history of solar and radiative instrumentation as a centenary of the first major experimental data gathering where some of the text are probably new to the web.
Do not take this data as accurate, it is considered defective. That said it will only be when the reader has much more information that they can make their own call.
So far as I know this data was collected by “Dr C G Abbot, Director of the Astrophysical Observatory of the Smithsonian Institution” who was a protege of Langley of the Langley unit of radiance fame. Abbot’s life work revolved around determining the solar constant but he had many assistants.
The instrument stands several feet high. The defect was the aperture allowing too much of the sky radiation surrounding the sun to enter, leading to slightly high readings (few %), later compensated. Are the figures I am using so compensated? Don’t know but since the original problem was long known it is very likely. Operating instructions can be posted if this is of interest.
His field work used the silver disc pyrheliometer which worked on a calorific principle, rate of change of the temperature of a chunk of metal irradiated by the sun. Unfortunately this was not a particularly good instrument.
Also involved where the three Swedish physicist Angstrom’s about whom my main tale will tell. Anders father of Knut father of Anders, a lot of room for confusion. The unit the Angstrom is in honour of the eldest.
Knut was an instrument maker as all good scientists need to be, contest as you wish. He invented the electrical pyrheliostat and pyrgeometer which measures “nocturnal radiation”, the latter apparently named by his son, the last of the three and main players.
The electrical compensation instruments were excellent and are essentially still in use today out in space although acknowledging the origination seems rarely done.
Figure 1 is the initial result of decoding a file set for which I had to write processing software. This is crude but good enough for a first look. There is far too much detail to explain here.
Two methods were used to determine solar constant
- long method
- short method
Initially Abbot used to long method, taken over some hours which included a solar spectrogram. Data of this kind is provided but is supposed to be not very good. Angstrom increasingly disputed the constancy over 3 or 4 hours and the short method eventually prevailed.
The short method, used for data shown here, deduces atmospheric absorption instead of from the transmission vs. frequency but from atmospheric depth, humidity, pressure, annulus pyranemeter. The details of what and how exactly has not come my way, are still missing jigsaw pieces.
I will reproduce part of abbot.txt here
The Smithsonian Astrophysical Observatory SOLAR CONSTANT Database
The Smithsonian Astrophysical Observatory (APO) gathered solar constant data during at least 49 years of solar monitoring. The solar constant is the total amount of energy received from the sun per unit time per unit area exposed normally to the Sun’s rays at the average Sun Earth distance and outside of the Earth’s atmosphere. The purpose of this APO project was to determine an accurate value for this energy flux and to determine whether or not the Sun’s total energy output is indeed constant in time.
Variability in the solar constant would have profound effects on terrestrial climate. Some of the phenomenon which are possibly related to solar flux variations are the advance and retreat of glaciers, occurrence of droughts, water levels in rivers and lakes, and changes in meridional sea level pressure. Whether or not the APO data does in fact show any variation in solar output has been a subject of considerable controversy. Many investigators claimed the measurements revealed real solar variations. Others felt the measurements gave no such proof.
One study of many done to correlate solar constant variation and solar activity used the monthly mean “long” method solar constant values measured at all stations and the monthly mean Wolf sunspot number. The linear least squares fit gives the value of the solar constant as S = [(1.94506 +/- .00059) + (.000008 +/- 000008)R], where S is the solar constant and R is the sunspot number.
Two basic measurement techniques were used, the “long” method and the “short” method. The “long” method utilized an instrument called a pyrheliometer that measured directly the total irradiance of the Sun. Combining this with a spectrobolometer which photographically recorded the spectrum of the of the
Sun, absolute values of the spectral irradiance could be deduced. Taking absolute irradiance at several air mass values (zenith angles) an extrapolation to zero air mass could be made. The extraterrestrial solar spectrum and hence solar constant were thereby determined.
The “short” method of monitoring the solar constant required considerably less time to compute and eliminated problems of varying atmospheric transmission. A pyranometer (aureole brightness meter) measures the brightness of the sky in a doughnut shaped ring about the Sun. This method relied on a “function” of precipitable water and pyranometer measurements to give a value for atmospheric transmission.
The validity of the data prior to 1923 was questioned by many, including the APO director from 1906 to 1944, Charles Greeley Abbot. Problems ranging from volcanoes to the invention of the improved measuring devices made this part of the data set of limited use. The remaining 31 years of data have been the
subject of a controversy of interpretation as mentioned earlier. Abbot and his APO staff claimed that the data showed a 0.2% increase in solar output from 1923 to 1954. Others argued that unconventional data manipulation made such a conclusion erroneous. Notwithstanding all of the short comings and controversy inherent in the data, this program is the longest and most carefully conducted solar radiation program made so far this century.
Out of curiosity I took slope and intercept, extrapolated linearly to 2010, figure is 1362.6 Watts sqm. Reading much into this is unwise given the know very complex situation. I have papers about the known errors. The scales used have been altered twice.
- conversion factor used 1Ly = 697.32 Watts/sqm
A version of this data collection
Or follow web link from here
“5b.) Charles Greeley Abbot solar constant database — Note: 2 years of scientific investigation are needed to bring this database into a scientifically useable research database”
I’m sure readers can dig out much more so as usual I am indicating a starting point.