Back to INAG Homepage Back to UAG-104 Contents page

VERTICAL-INCIDENCE SOUNDING DATABASE AND ITS PRODUCTS

T. L. Gulyaeva

Institute of Terrestrial Magnetism, Ionosphere and Radio Wave Propagation, Academy of Sciences,142092, Troitsk, Moscow Region, Russia

Abstract

Long-term VI sounding observations present a unique database suitable for determining quiet and disturbed ionospheric conditions. Based on proposed criteria, local and regional catalogues of ionospherically disturbed and quiet days are developed from an analysis of vertical-incidence sounding data over the European area during 1964-1990. Comparison of ionospherically quiet days with geomagnetic quiet periods for 1976-1988 show coincidence for no more than 30% of the days considered. The conventional practice of defining the state of the ionosphere by comparison with geomagnetic conditions should be replaced by specification of the proper ionospheric quiet and disturbed periods.

Satellite and rocket borne observations of the ionosphere yield a mix of quiet and disturbed ionospheric parameters. Similarly the short-term series of incoherent scatter observations don't allow results to be sorted by their degree of disturbance. At the same time there are long-term ground-based vertical-incidence sounding observations which provide a unique opportunity to separate the above conditions. Needless to say the development of ionospheric models requires clear specification of quiet conditions, with the degree of disturbance quantified by a proper system of indices.

At the URSI-COST238 (PRIME) Workshop on Verification of Ionospheric Models and maps at Roquetes, Spain, May 1992, there were no publications, reports or presentations on criteria for defining quiet days in the ionosphere [Hanbaba, 1993]. Nevertheless there is an urgent need to clear up this problem since most available models and maps are built up just for quiet conditions. Specification of conditions in the ionosphere is often made by reference to geomagnetic quiet or disturbed conditions. To what extent are available ionospheric models and maps valid in the real ionosphere? How should ionospheric conditions be ranked to show the degree of disturbance and quietness? Which steps should be taken first in analysis of the ionospheric data? These and other items need to be resolved immediately.

In response to that urgent need, work projects have been planned and cataloguing of ionospherically quiet and disturbed conditions has begun. During the year which has passed since the Roquetes meeting, criteria were developed and the first catalogues produced for ionospheric quiet and disturbed days [Cooper et al, 1993]. A general statistical approach for determining the weighted skewness of the histogram was introduced, to obtain the occurrence frequency of deviations of the current values of a specified parameter from its median or other background level [Gulyaeva, 1993a,b]. A mean-weighted measure of the histogram distortion, with an increasing contribution of the more disturbed sections of the histogram, serves as a measure of the disturbance of the particular parameter.

We have applied the proposed criteria to the long-term database of the VI sounding ionospheric observations currently available on optical disks [Allen, 1991; Hanbaba, 1993], to produce a second-generation database of ionosphere disturbance indices. As a result we obtained hourly, semi-diurnal, diurnal, monthly and annual indices of the negative Dm- and positive Dm+ ionospheric disturbances. Each annual collection of results, for each ionospheric station, includes 10 files originating from a set of 12 monthly tables of ionospheric characteristics in URSI format (routinely presenting parameter MUF3000F2, re-evaluated from two initial tables of critical frequency foF2 and the height equivalent M3000F2; or missing the latter the initial table of foF2 is used). From the initial tables the hourly values of indices of deviations from the monthly median are calculated. Note that the weights are increased when descriptive or qualifying letters are provided with the numerical hourly value, and missed observations are taken into account as disturbed conditions. So the hourly disturbance indices of the negative and positive deviations, Dm- and Dm+, are obtained. Similarly the histograms of frequency of occurrence of different hourly Dm- and Dm+ indices are analysed with various temporal filters (semi-diurnal to annual scales).

 

TABLE 1 Part of the catalogue of 5 ionospherically disturbed days (D1 = the most disturbed day) and 10 quiet days (Q01 = the quietest day) at Moscow (MO) and Tokyo-Kokobunji (TK) during the 3d VIM period, for 4 seasons of 1988-1992.

 

STYRMN

D1

D2

D3

D4

D5

Q10

Q09

Q08

Q07

Q06

Q05

Q04

Q03

Q02

Q01

 

MO9111

09

08

10

22

02

15

03

19

29

21

23

25

17

26

12

FALL

MO9112

18

30

05

24

23

21

22

12=

15

10

20

17

31

06

19

WNT

MO9203

29

01

30

05

25

06

07=

12

02

16

17

03

08

10

15

SPR

MO8807

11

22

23

27

21

20

07=

28

14

17=

26

16

04

15

24

SMR

TK9111

09

02

19

05

24

22

03

28

29

16

10

14

04

07

11

FALL

TK9112

1

05

23

28

30

15

10

11

31

12

07

26

09

03

22

WNT

TK9203

18

07

30

06

29

26

01

04

11

02

22

10

27

05

19

SPR

TK8807

01=

02

20

11

04

05

08

15

29

31

24

18

27

13

30

SMR

Specific criteria for obtaining a catalogue of quiet and disturbed days are proposed [Gulyaeva, 1993c]. Three signatures of disturbance are taken into account (diurnal indices Dm- for negative deviations from the median, diurnal indices Dm+ for positive ionospheric disturbances with ionisation input into the F region, and their range Rm for each day of the month). These indices are analogous to the well known three types of auroral electrojet indices - the lower hourly value Al, the upper value Au, and the range AE-index [Mayaud, 1980]. To define 5 ionospheric disturbed days and 10 quiet days per month, similar to those obtained from the planetary geomagnetic Kp index, the diurnal values of Dm-, Dm+ and Rm are ranked in decreasing order and the sum of current ranks is calculated for each day of the month. The most disturbed day is identified by the greatest sum of current ranks (as used to judge sport competitions), and the quietest day has the least sum of current ranks. An example of these results is presented in Table 1 for the third VIM period [Reinisch, 1993].

Ranking by 5 disturbed days and 10 quiet days is carried out independent of the mean monthly or annual level of disturbance within the solar cycle. But those important factors are crucial for a study of the consistency of parameters for different geophysical fields. Evidence of different levels of disturbance at Tokyo (Kokobunji) and Moscow for the third VIM period (see Table 1) is presented in Figure 1, where histograms show the frequency of occurrence of the hourly indices Dm- and Dm+ for 4 reasons. For convenience of computer analysis we can denote every day as quiet (0) or disturbed (1), separating these daily Dm indices with floating thresholds equal to the monthly index of Dmm- or Dmm+ provided with Figure 2. This is an example of a Bartels diagram for ionospheric semi-diurnal negative and positive disturbances, from the monthly mean levels at Moscow during 1991, which could be compared with a simple voting procedure with relevant diagrams for the geomagnetic indices.

Further selection of really quiet days is made within one standard deviation from Dm=1 (the least value of Dm by definition) among those days when diurnal Dm- and Dm+ do not exceed the monthly threshold. The results of such a selection of Q and D days (0 or 1) for each month of 1991 at Moscow are presented in

TABLE 2. Disturbed (1) and quiet (0) days in the ionosphere at Moscow during 1991. CURRENTLY THIS TABLE DOES NOT REPRODUCE WELL FOR SOME BROWSERS. THE PROBLEM IS BEING ADDRESSED.

 

STYRMN _

 

2

_

4

_

6

_

8

_

0

_

2

_

4

_

6

_

8

_

0

_

2

_

4

_

6

_

8

_

0

_ Q-DAYS

MO9101 1

1

1

1

0

0

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

0

1

1

1

1

1

1

1

1

1 3

MO9102 1

1

1

1

1

0

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

 

 

1

MO9103 0

0

1

1

0

1

1

1

1

1

1

0

1

0

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1 5

MO9104 1

1

1

1

1

1

0

1

1

0

1

1

0

1

0

1

1

1

1

1

1

1

1

1

0

1

1

1

1

1

5

MO9105 1

1

1

0

0

1

0

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

0

1

1

1

1

1

1

0

1 5

MO9106 1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

0

1

1

1

1

0

0

1

1

1

1

3

MO9107 1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

0

1

1

1

1

1

1

1

1

1

1

1

0

1

1

0 3

MO9108 1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

0

1

1

1

1

0

1

1

1

1

1

1

1

1

1 2

MO9109 1

0

1

1

0

1

1

1

1

1

1

0

0

1

1

1

1

1

0

1

1

1

1

1

1

1

1

1

1

1

5

MO9110 1

1

1

1

1

1

1

1

1

0

0

0

1

1

1

1

1

1

1

1

1

0

1

1

1

1

1

1

1

1

1 4

MO9111 1

1

1

1

1

1

1

1

1

1

1

0

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

0

1

2

MO9112 1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1 0

 

 

Month

Jan

Feb

Mar

Apr

May

June

July

Aug

Sep

Oct

Nov

Dec

YR

L3MO91 Dm-

19

21

25

25

25

27

26

24

21

30

24

20

239

L3MO91 Dm+

20

18

20

19

21

23

20

23

20

24

25

21

211

 

TABLE 3 Ionospherically quiet days coincident with part of the 7-day geomagnetically quiet periods during 1976-1988.

 

Year

 

Month

(a) Gulyaeva's Catalogue

(b) Stanislawska's Catalogue.

1976

May

 

12,18

 

Jul.

19,20

18,19

 

Aug

11,12,19

12,13,15,18,19

 

Oct

21

21

 

Nov

4

5,6

1977

Jun.

7

4

 

Nov

21,22,23

19

1978

Jul

29,31

 

 

Dec

 

6

1980

Jan

18,23

 

 

Mar.

11,13,15,16,17

8,11,13,15,17,18

1985

Jun.

 

13,14

 

Aug

 

3,4,5,7,9

1986

Jan

14

11,12,14,19

 

May

12,13,14,15

9,10,11,12,13,14,15

 

Jul

6,7,12,13,20

4,5,12

 

Dec

9 3,4,7,8,9

 

1987

Apr

 

29,30

 

May

15,20,21

1,2,4,5,6; 20

 

Dec

30

27,28,30

1988

Jan

28,29

28,29

 

Mar

 

20,22,23,24

 

Sep

10

4,10

 

Nov

19

19,22,23,24,25

Table 2, as part of a local catalogue of ionospheric disturbed and quiet days. Note that while 0 means a really quiet day, 1 implies both moderate and heavy disturbances. To distinguish further the degree of disturbance one should look at the hourly Dm- and Dm+ indices. Similar results have been obtained using data from European and other world-wide ionospheric stations, comprising so far about 10% of the available planetary archive of VI sounding data.

The present average European Catalogue of ionospheric disturbed (1) and quiet (0) days is compared in Table 3 with an Ionospheric Catalogue for the same area [Stanislawska, 1993] and with the geomagnetic quiet 7-days periods [from Joselyn, 1990]. We have 39 Q-days by our criteria, 71 Q-days by Stanislawska, giving 21 common Q-days in the two ionospheric Catalogues, against 225 geomagnetic quiet days for the same periods. Thus, only 10 to 30% of quiet days are common in the ionosphere and the geomagnetic field simultaneously. This means that the conventional rules for analysing ionospheric data based on available geomagnetic indices require re-examination. We need criteria and catalogues of ionospheric quiet, moderately disturbed and heavily disturbed conditions, to permit analysis of ionospheric data based on the ionosphere disturbance and quietness. To speed up and broaden such studies we have published the main subprogram for producing hourly, monthly and annual ionospheric indices [Gulyaeva, 1993b]. Exchange of results from such analyses through the relevant projects of URSI and other international bodies would stimulate progress in ionosphere research.

Acknowledgments

Presentation of this paper has been made possible thanks to Travel Grant No. 0412/2 provided to Dr T. L. Gulyaeva by the International Science Foundation (USA).

References

Allen J. (1991), Selected geomagnetic and other solar-terrestrial physics data of NOAA and NASA, CD-ROM, WDC-A-STP, Boulder, Co., USA,

Cooper J., K. K.Barbatsi, T. L. Gulyaeva, G. A. Moraitis, T. A. Th. Spoelstra, I. Stanislawska, S. M. Radicella, M. -L. Zhang (1993). PRIME Catalogue of undisturbed days No. 1 (April 1993). In: Proc. of PRIME COST 238 Workshop, University of Graz, Austria, 145-149.

Gulyaeva T. L. (1993a), Voting procedure for distinction of geomagnetic quiet and disturbed conditions at the ionosphere data analysis and modelling, in: Proceedings of the PRIME/URSI joint Workshop on "Data validation of ionospheric models and maps (VIM)", Roquetes, Spain, May 1992, Memoria No.16, Ed. by L.Alberca, pp.352-367.

Gulyaeva T. L. (1993b), Development of database of the ionosphere disturbances, Adv. Space. Res., in press (1993)

Gulyaeva T. L. (1993c). Criteria for the definition of the quiet ionosphere. In: Proc. of PRIME COST 238 Workshop, University of Graz, Austria, 129-136.

Hanbaba R. (1993), Catalogue No.3 of European ionosonde data in the COST238 databank, in: Proceedings of the PRIME/URSI joint Workshop on "Data validation of ionospheric models and maps (VIM)", Roquetes, Spain, May 1992, Memoria No.16, Ed. by L.Alberca, pp.305-324.

Joselyn J. A. (1990). Forecasting magnetically quiet periods.In: Solar Terr. Predictions: Proc. of Leura-1989 Workshop, Australia. Ed. R.J.Thompson et. al., NOAA, Boulder, CO, USA, pp. 102-107.

Mayaud P. (1980), Derivation, meaning and use of geomagnetic indices, Geophys. Monogr. Series 22, AGU, Washington, D.C., USA.

Reinisch B. W. (1993), Chairman's Report, in: Proceedings of the PRIME/URSI joint Workshop on "Data Validation of ionospheric models and maps (VIM)", Roquetes, Spain, May 1992, Memoria No.16, Ed. by L.Alberca, pp.13-16.

Stanislawska I. (1993), Catalogue No.2 of European quiet ionospheric days 1966-1991. In: Proc. of PRIME COST 238 Workshop, University of Graz, Austria, 137-143.

MN:

01

02

03

04

05

06

07

08

09

10

11

12

YR

L3MO91 Dm-

19

21

25

25

25

27

26

24

21

30

24

20

239

L3MO91 Dm+

20

18

20

19

21

23

20

23

20

24

25

21

211

Back to INAG Homepage Back to UAG-104 Contents page