An improved algorithm for determining neutral winds from the height of the F₂ peak electron density

Abstract

One of the greatest uncertainties in modeling the ionospheric densities and temperatures lies in the neutral winds that affect the vertical ion drift. Wind measurements are difficult to make, and current wind models are not adequate. An alternative is to deduce an equivalent meridlonal neutral wind from measurements of the height of the F2 peak electron density. This method has been effective in allowing ionospheric modelers to reproduce the observed height with reasonable precision when the peak height does not vary rapidly. However, when there are rapid movements of the peak height, this method underestimates the magnitude of the necessary wind changes, and it is also complicated and computer intensive to do accurately. This paper introduces an improved algorithm for use in time dependent ionospheric models that reduces the complexity of the procedure while more accurately reproducing the observed heights. The improvement is achieved by estimating the wind that is needed to reproduce the observed height at the next time step using the calculated height and wind at the current time step. Thus the winds are continuously being a•usted to bring the calculated height into better agreement with the observed height during the time dependent simulation. The winds from the new algorithm agree well with both optical and radar measurements made at Arecibo on August 17-18, 1982. 1.density (nmF9.) and the height at which the peak density occurs (hmF•). Normally, the wind is poleward during the day, enhancing the downward diffusion to a region of increased loss and thereby reducing the peak density (nmF2) and lowering the height of the layer. At night the wind becomes equatorward, inhibiting the downward diffusion, raising the height of the layer, and helping maintain the nighttime F• layer density. Thermospheric neutral winds have been measured from satellites with mass spectrometers [Spencer et al., 1981] and by Fabry-Perot spectrometers [Hays et al., 1981]. Currently, no satellite is making wind measurements. Ground-based measurements continue to be made with both Fabry-Perot instruments and incoherent scatter radars. However, neither of these methods can supply wind measurements on the global scale needed to model the ionosphere. The optical measurements are linfited by such factors as clouds, daylight, and the phase of the Moon. The radar measurements require expensive equipment and are often unavailable for routine ionospheric measurements. The lack of suitable measurements on a global scale makes it difficult to accurately model the ionospheric electron density. This in turn makes it difficult to accurately model the many emissions, densities, and temperatures that are sensitive to electron density. Richards and Tort [1986] to suggest using the height of the peak F•. region electron density (h,,.F2) to deduce neutral winds. The relationship between h,,,F2 and the meridional wind has been shown to be approximately linear for small winds under steady state conditions 17,839 17,840 RICHARDS: WINDS FROM HEIGHT OF F2 PEAK ...

Subauroral red arcs as a conjugate phenomenon: comparison of OV1-10 satellite data with numerical calculations

“…(A13 , A14) given by Pavlov (1994). The large uncertainities in the calculated h lead to large errors in the modeled electron densities and peak height altitude, hmF2, of the F2-layer (Richards, 1991;Pavlov, 1994). This problem was largely overcome when Richards (1991) developed a technique for deriving equivalent neutral winds from the observed hmF2.…”

“…The large uncertainities in the calculated h lead to large errors in the modeled electron densities and peak height altitude, hmF2, of the F2-layer (Richards, 1991;Pavlov, 1994). This problem was largely overcome when Richards (1991) developed a technique for deriving equivalent neutral winds from the observed hmF2. Ideally, the observed hmF2 should be available for several hours before the time of the comparison, but very often only a limited set of hmF2 measurements is available or we have not got the observed hmF2 and thus cannot use the method developed by Richards (1991) or our modi®cation of this method (Pavlov and Buonsanto, 1997) to calculate an equivalent plasma drift velocity.…”

“…This problem was largely overcome when Richards (1991) developed a technique for deriving equivalent neutral winds from the observed hmF2. Ideally, the observed hmF2 should be available for several hours before the time of the comparison, but very often only a limited set of hmF2 measurements is available or we have not got the observed hmF2 and thus cannot use the method developed by Richards (1991) or our modi®cation of this method (Pavlov and Buonsanto, 1997) to calculate an equivalent plasma drift velocity. To overcome this problem, we can try to use hmF2 IRI given by the IRI-90 model (Bilitza, 1990) to calculate an equivalent plasma drift velocity, IRI , for some hypothetical undisturbed conditions with the same solar and time conditions as during the studied conditions:…”

“…In this study we employ model A and compare results with those from models B and C to evaluate the e ects of the uncertainties in these rate coe cients on calculated integral airglow intensities at 630 nm in the SAR-arc regions. We have no information on hmF2 and thus can use neither the method developed by Richards (1991) nor our modi®-cation of it (Pavlov and Buonsanto, 1997) to calculate an equivalent plasma drift velocity. To overcome this problem, we use the approach of Eqs.…”

“…During quiet conditions the modeled hmF2 are very close to hmF2 IRI . If we have the observed peak height altitude, hmF2 exp , of the F2-layer in the northern or southern hemispheres then the algorithm developed by Richards (1991) with our modi®cation (Pavlov and Buonsanto, 1997) is used to calculate d for this hemisphere…”