THE NEAR-EARTH MAGNETIC-FIELD FROM MAGNETOSPHERIC AND QUIET-DAY IONOSPHERIC SOURCES AND HOW IT IS MODELED

Early attempts to model fields from sources in the magnetosphere durin g magnetically quiet times were unsuccessful, This is attributed to th e presence of fields from lithospheric sources and to deficiencies in the data distribution. The availability of data from the Magsat satell ite allowed definitive determination of the first-degree field from ma gnetospheric sources. Temporal variation of those fields on time scale s of hours to a few years was implemented by use of a proxy function t hought to mimic that variation. The Dst index was chosen as the proxy function, Extension of determination of the first-degree magnetospheri c field to earlier times, namely to 1900, is accomplished first by use of the annual average of the aa index as a proxy function and then by use of cubic B-splines as temporal basis functions. The model first-d egree field is well correlated with the annual sunspot number, both in its general level changes and in its detailed year-to-year variation. The correlation coefficient of the year-to-year variation of the firs t-degree, zero order, external term with the annual sunspot number is 0.77 when the external term lags the sunspot number by 2 years. Anothe r extension of the results made possible by the Magsat data is simulta neous determination of a model of the fields from the magnetosphere, t he ionosphere, and internal to the Earth. In this case, data from the Magsat and POGO satellites and magnetic observatory hourly mean and an nual mean data were analyzed together. Though still in an experimental stage, these models are shown to fit all the data closely and to be a ble to readily isolate the fields according as the source is in the ma gnetosphere, ionosphere, or Earth.