AN EQUATORIAL ZONAL ION DRIFT MODEL FOR JICAMARCA

1
AN EQUATORIAL ZONAL ION DRIFT MODEL FOR
JICAMARCA 1Jonas R. Souza, 2B. G. Fejer, 3A. S.
Santos, 4A. E. Costa Pereira, 5Dieter Bilitza and
1M. A. Abdu 1Instituto Nacional de Pesquisas
Espaciais, INPE, Brazil 2Center for Atmospheric
and Space Science, Utah State University, Logan,
Utah, USA 3Federal University of Goiás, Catalão,
Goiás, Brazil 4Universidade Federal de
Uberlândia, Uberlândia, Minas Gerais,
Brazil 5GSFC, Code 612.4/Raytheon, Greenbelt, MD
20771, USA Abstract. We have published a
regional empirical model for the disturbed- and
quiet-time equatorial zonal plasma drift near the
F region peak. Incoherent scatter radar database
from Jicamarca Radio Observatory registered
between 1970 and 2003 and Bernstein polynomials
as base functions were used to construct this
model which will be incorporated into the IRI.
The model also includes dependences with solar
activity and season. Our quiet-time model results
confirm that the daytime drifts are westward and
are nearly season and solar cycle independent.
The nighttime drifts are eastward, have larger
magnitudes, and increase strongly with solar
flux, particularly near equinox and December
solstice. Enhanced geomagnetic activity drives
small eastward perturbation drifts during the day
and much larger westward disturbance drifts at
night. The nighttime drift perturbations are
largest near midnight and increase strongly with
solar flux near equinox and December solstice but
are essentially absent near June solstice.
2
Jicamarca Geog. 11º57S,76º52 W Mag. Dip
2oN
Map showing Jicamarca station and others
locations with Incoherent Scatter Radars.
3

• DATA ORGANIZATION
• We have used 206 days of measurements from
  1970 to 2003. The data correspond to averages
  near F region peak (between 300-500 km) and were
  grouped into the following bins
• Quiet time zonal drifts
• 1. Drift data were binned in half an hour
  intervals centered 15
• min.
• 2. Seasonal bins were defined by June
  solstice (May-August),
• December solstice (November-February)
  and Equinox
• (March.-April September-October)
• 3. Quiet time condition Kp ? 3.
• Disturbed zonal drifts
• We first determined the perturbation drifts
  by subtracting from each half-an hour averaged
  drift the corresponding season and solar cycle
  dependent average quiet time value (?Zonal drift
  versus ?Kp). To account for solar cycle effects,
  we have studied separately the perturbations
  drifts corresponding to decimetric solar flux
  indices smaller or equal to 140, and larger than
  120.

4
MODEL
DESCRIPTION The zonal drift is modeled using
Bernstein polynomials as base functions. The
general form of our model is given by where
ak are the coefficients and B0,n, Bn,n is a
Bernstein basis of degree n defined as We
define chi-square as and ak are determined
by minimizing chi-square Bernstein polynomial
degrees used in the modeling 1 for Kp
dependence, 2 for latitudinal and solar flux
dependence, 11 for local time dependence. Linear
interpolation was used to have zonal drift
between 2 seasons. (See Fejer et al., 2005
JGR,110, A12310)
5
RESULTS model validation - Quiet time -
Figure 1. Seasonal and solar cycle dependence of
average F region quiet-time zonal drifts
overJicamarca. The scatter bars denote the
standard deviations and the solid curves
represent the model results (See Fejer et al.,
2005 JGR,110, A12310).
6
RESULTS model validation - Disturbed time -
Figure 2. Variation of zonal perturbation drifts
with magnetic activity near midnight and noon.
Here ?Kp (3-15) 0 corresponds to an average Kp
1.8 over the preceding 3 to 15 hours (See Fejer
et al., 2005 JGR,110, A12310).
7
RESULTS model validation - Disturbed time -
Figure 3. Comparisons of modeled and binned zonal
disturbance drifts.
8
MODELED RESULTS
Figure 4. Seasonal and solar cycle dependence of
quiet and disturbed zonal drifts.
9
MAIN CONCLUSIONS
We have developed an empirical model for
prediction of equatorial zonal drift velocities
using Bernstein polynomials. This method has
produced zonal drift variation patterns on
different time scales showing excellent
resemblance with observations. Our quiet time
Jicamarca zonal drifts are in good agreement with
results from previous radar studies and other
measurement techniques. We have examined in
detail, for the first time, the response of
enhanced geomagnetic activity on the Jicamarca
zonal drifts. Our results indicate that,
surprisingly, geomagnetic activity effects on the
zonal drifts are strongly season and solar flux
dependent. The Jicamarca zonal disturbance
drifts occur predominantly at night, are largest
during equinox and December solstice high solar
flux conditions, and are essentially negligible
near June solstice. The daytime perturbations are
eastward, have small magnitudes, and occur about
3-15 hours after enhanced geomagnetic activity.
The nighttime disturbance drifts are westward
with peak values at about 0300-0400 during low
solar flux periods. They increase with solar flux
and reach largest values near midnight high solar
flux periods.