3. Morphological Results

Figure 1

Figure 2

[4]  In the beginning we present some examples of Q disturbances to get an idea of how they look in the equatorial F2  layer. According to our analysis (see later) both negative and positive Q disturbances are the most frequent in the dark LT sector and in the daytime sector (only positive disturbances). Therefore three examples of Q disturbances observed at Huancayo are given in Figures 1 and 2. A strong nighttime negative disturbance on 12 August 1962 is shown in Figures 1a and 1b. Note that only nighttime period was subjected to the N_mF2 decrease, while N_mF2 values for the whole previous daytime hours of 11/12 August coincide with the median. Similar effect with less amplitude took place on the previous night of 11 April. In both cases a preceding decrease is seen in the h_mF2 variations calculated using the expression by Bradley and Dudeney [1973]. That was a strong disturbance with N_mF2 decrease by a factor of 6 and h_mF2 decrease by 30-40 km. The period was characterized by very low geomagnetic activity.

[5]  The 22/23 May 1965 period presents a case of strong positive nighttime Q disturbance (Figure 1d). Again the disturbance has developed under very quiet geomagnetic conditions and took place for two adjacent nights, while daytime N_mF2 values were close to the median. By analogy with the previous case the positive N_mF2 disturbance was preceded by positive deviations in h_mF2. The positive N_mF2 effect in this case was a factor of 4 and about 40 km in h_mF2. Therefore nighttime equatorial F2 -layer Q disturbances demonstrate synchronous (of one sign) N_mF2 and h_mF2 deviations and this is different from midlatitude Q disturbances (MDL). These examples show that equatorial quiet time disturbances may be very large with the amplitude comparable to normal F2 -layer disturbances related to geomagnetic activity.

[6]  Figure 2 gives an example of daytime positive Q disturbance observed on 16 February 1966. The daytime positive effect is not so impressive (only a factor of 1.7 in N_mF2 ) compared to the nighttime ones, but this implies pretty large changes in aeronomic parameters. The daytime F2 region is mainly controlled by local processes, while in case of nighttime disturbances we have a cumulative effect resulting in large N_mF2 deviations by the end of night (Figure 1). Contrary to the nighttime case, here we have antiphase N_mF2 and h_mF2 variations. This is also different from the midlatitude daytime Q disturbance case (MDL). It should be also stressed that this differs from the usual midlatitude daytime positive storm effect when N_mF2 and h_mF2 vary synchronously. All these peculiarities are due to the equatorial F2 -layer formation mechanism, but this will be discussed in this paper only at a qualitative level. Model calculations along with a quantitative analysis of various aeronomic parameters contribution will be given later elsewhere. Let us consider morphological results obtained over the periods of observations available at the two stations.

3.1. Duration and Total Number of Disturbances

Figure 3

[7]  Distributions for the occurrence of positive and negative Q disturbances versus their duration are shown in Figure 3 for both stations. Three levels of solar activity are shown separately although the total number of events is small for solar maximum conditions. Similar to middle latitudes (MDL) short-term ( <3 hours) deviations are seen to be the most numerous and they may be attributed to short-term ionosphere fluctuations which lie beyond our scope. We are interested in longer disturbances which can be related to background changes in the controlling aeronomic parameters. A 3-hour (4 hourly successive N_mF2 values) threshold was accepted for our analysis. Contrary to middle latitudes, negative Q disturbances are seen to be more numerous than positive ones at both stations and all levels of solar activity. The distributions are seen to be broader at Huancayo, that is the percentage of long (both negative and positive) disturbances is larger in the American sector. In general, long-duration disturbances are more numerous at solar minimum compared to solar maximum. The occurrence of long deviations is low during solar maximum especially at Kodaikanal. Therefore, in some cases we had to put together all solar activities to present the results.

3.2. Occurrence Versus Local Time

Figure 4

[8]  Distributions of the occurrence for negative and positive Q disturbances (duration 3 hours) versus local time (LT) are given in Figure 4. It was possible to show separately three solar activities for negative disturbances at Huancayo and this is important for further discussion of the mechanism. In general, both stations demonstrate similar occurrence distributions, but some differences are also obvious. There is a well-pronounced night-early morning peak in the occurrence frequency for negative disturbances, while they are practically absent during daytime hours. This nighttime peak is broad at Kodaikanal overlapping the whole dark period with the maximal occurrence in the premidnight LT sector. At Huancayo the nighttime peak is large and narrow maximizing in the early morning LT sector. There is a pronounced dependence on solar activity level (Figure 4a) with the majority of negative disturbances clustering in the early morning LT sector at solar maximum, while the disturbances are spread over the whole dark LT sector under solar minimum. The disturbances start to occur early in the evening at solar minimum, but they are practically absent in the evening sector at solar maximum.

[9]  Positive Q disturbances exhibit two maxima, the nighttime-early morning and the daytime ones. Similar to the negative disturbance case the variation at Kodaikanal is shifted to earlier LT hours compared to Huancayo. The nighttime peak is broad at Kodaikanal and covers the whole dusk-dawn LT sector. At Huancayo the peak localizes in the postmidnight LT sector similar to the negative disturbance case. In general, the equatorial Q disturbance pattern looks simpler compared to the midlatitude one (MDL) and this may help understand the revealed morphology.

3.3. Annual Variations

Figure 5

[10]  The season/solar activity distribution of the number of cases for negative and positive Q disturbances at the two stations is given in Tables 1, 2, and 3. The summer and winter seasons include the months selected in accordance with the different hemisphere location of the two stations (November, December, January, February, winter/summer; May, June, July, August, summer/winter; and March, April, September, October, equinoxes). For the presentation obviousness the annual distribution of occurrences (all solar activities put together) are given in Figure 5. The daytime LT sector for positive disturbances is shown only for Huancayo due to insufficient number of events at Kodaikanal (Table 3). In general, seasonal variations are more distinct and pronounced at Huancayo. The occurrence of negative disturbances at Huancayo exhibits a broad and well-pronounced distribution with the peak maximizing in winter (Figure 5a). This is valid for all solar activities (Table 1). The morphology of positive nighttime Q disturbances at Huancayo is similar to the negative one. They also show a well-pronounced seasonal variation of the occurrence maximizing in winter (Figure 5c) under all solar activities (Table 2). The disturbances are also clustering in the early morning LT sector (Figure 4). An interesting seasonal distribution exhibit daytime positive Q disturbances at Huancayo (Figure 5e), which is inverse to the distribution of nighttime disturbances. Although the total number of events is small (Table 2), the seasonal variation is well pronounced with the maximal occurrence in summer while these disturbances are completely absent in winter (May-August).

[11]  At Kodaikanal the morphological pattern is quite different. Figures 5b and 5d and Tables 1 and 3 show that seasonal variations are not distinct, although there is a tendency for a distribution with winter and summer maxima. As the equatorial F2 region is strongly controlled by vertical EB drifts, the revealed morphological differences tell about longitudinal and perhaps hemispheric differences in vertical drifts at the two stations.

3.4. Amplitude of Disturbances

Figure 6

[12]  The N_mF2 _obs/N_mF2 _med ratios for negative and positive Q disturbances versus local time are given in Figure 6 for Huancayo. Similar results were obtained for Kodaikanal, but they being less representative are not given here. Here all disturbances with the duration 1  h were taken into account and the maximal ratio observed within each particular disturbance along with the corresponding LT moment were used to draw the plots. All durations were considered for the following reason. The disturbances practically do not commence during the sunrise and for a couple of hours after (Figure 4a and 4b); therefore, considering only long-duration disturbances, we are loosing many of them which commence within 1-2 hours before the dawn and then being terminated by the sunrise. Therefore we have included all disturbances to have a picture more complete. Median values over the observed ratios were found for each LT to demonstrate average diurnal variations.

[13]  Negative Q disturbances exhibit a pronounced dependence on LT and solar activity level. The disturbance effect strongly increases from daytime hours ( N_mF2 _obs/N_mF2 _med ratio 0.6 under all solar activity levels) to nighttime hours, the effect being stronger at solar minimum when the N_mF2 decrease may reach an order of the magnitude and even larger. The disturbances are more moderate under high solar activity (Figure 6e). This dependence is seen not only in the individual disturbances but in the median values as well. The maximal negative effect is reached by the early morning hours (at 2000-0400 LT).

[14]  In case of positive Q disturbances both dependencies look more moderate. Similar to the previous case the maximal disturbance amplitudes are observed in the early morning LT sector. However, the amplitude of the majority of disturbances is less than a factor of 3 even at solar minimum (Figure 6b). The dependence on solar activity is not distinct for median values, although the tendency for median ratios to increase by the nighttime hours is seen.