The vertical drift may be calculated from ionograms as V=d(hF)/dt,
it is a valid representation of the true vertical drift whenever the F layer height was near and above 300km as demonstrated by Bittencourt and Abdu(1981). For height less than 300km d(hF)/dt represents an apparent vertical drift, which represents the sum of the true vertical drift and a contribution due to the recombination processes. Therefore, you can estimate the true vertical drift ,to do it is necessary to calculate the term of loss due to recombination, which is given by bH (Subbarao and Krishnamurthy,1983), where b is the recombination coefficient and H is the scale height of ionization.The recombination coefficient given by b=k1[N2]+
k2[O2] and the reaction coefficients k1 and k2 may be obtained from Anderson and Rusch(1980). See also the methodology of Nogueira et al., 2011 (doi:10.1016/j.jastp.2011.02.008)
thaks to Prof. Nogueira for his contribution. However, Bertoni et al. (2006) had used ionospheric data obtained from digisonde DPS-4 and an Incoherent Scatter Radar (ISR) measurement, both co-located at Jicamarca Radio Observatory, Peru, to compare vertical plasma drifts from the two instruments (DPS and ISR). The observations were recorded during the campaign periods of 7-11 October, 2002 and 19-23 March, 2003 at Jicamarca, which signifies equinox season. They reported that within 1700-2100h (representing sunset hours) and starting from around 0200-0300h through 0800h for h’F ≥ 300 km , both the ISR and Digisonde observations of vertical plasma drifts are relatively well comparable, but deviate away from each other at other hours.
The correlation coefficient (R) between the night-time Jicamarca ISR drift observation and the inferred drift obtained from hmF2 over Ilorin (e.g. Adebesin et al., 2013a) between 1700-2100h is 0.7. For 0200-0800h, R = 0.6. This correlation values are consistent with Bertoni et al. (2006) observation. Recall that the results of Adebesin et al. (2013a) had suggested that the general theory that vertical drifts obtained by digisonde measurements only match the E x B drift if the F layer is higher than 300 km is reliable, but may not hold for the night-time period within 2200-0600h and early morning (0600-0900h) periods under condition of solar minima (tentative). Hence, a good correlation relationship between the ISR and DPS-4 during conditions of h’F ≥ 300 km (Bertoni et al., 2006) and hmF2 ≤ 300 km (Adebesin et al., 2013a) between the hours of 1700-2100h and 0200-0800h should pose a question for future work.
I agree with the explanation by P. A. B. Nogueira as also with some parts of the other comments that appeared above. The main point to keep in mind is that the Digisonde measures the radio wave reflection height and the rate of change of the reflection height is a proxy for the vertical drift provided that other factors such as the ion production by ionizing radiation or loss by recombination do not influence the reflection height. Therefore during the day time and at heights below approximately 300 km where both these factors are dominant the vertical drift measured as dhF/dt, or from the Digisonde Doppler information, can be incorrect in varying degrees. The evening hours approaching the sunset and the night hours are the intervals when the dhF/dt or the Doppler data from the Digisonde can provide reliable values of the vertical drift provided the height at which the drift is measured is near and above 300km (where the recombination effect is negligible). Further as the height increases and the diffusion effect becomes dominant the vertical drift that is measured should be fast enough to avoid the effect of diffusion on the rate of change to the reflection height.
Thanks Prof. Abdu for your clarification. However, I have some few more questions. 1. Why is the chemical correction necessary when hF > 300 km? 2. Which is better between hmF2 and h'F to infer vertical drift from digisonde at daytime and at nighttime? 3. Can I get your article: Bittencourt and Abdu, 1981? I am presently working on a review of drifts obtained by both hmF2 and h'F at an African equatorial station. You can help send any other material of interest on drifts and zonal electric field at equatorial station. My mail is [email protected] or [email protected]
When the hF > 300 there is no need to make chemical correction.
The reason why h'F and hmF2 are more often often used for obtaining vertical drift is simply because they are standard parameters more easily accessible in the ionosonde data bank. It is more appropriate to use an intermediate height, preferably the true height. Any virtual height should also be okay since the time rate of change may not be very different between the true and virtual heights. I am sending you some reprints in your e-mail address.
Prof. Abdu, Thanks for the reprints. I have sent you some of mine too. The 1st last question is Why is the chemical correction necessary when hmF2 < 300 km?