The ALIS stations

The design and deployment of the ALIS stations involves a large number of considerations. Each station must be designed for unmanned remote-controlled operation during extended time-periods in low-population regions with a sub-arctic climate. The technical design of the ALIS-station, which constitutes a Ground-based Low-light Imaging Platform (GLIP), is considered in Appendix A.

The main scientific instrument at the ALIS-station is the ALIS-imager (covered in detail in Chapter 3). Although the stations are primarily designed for optical instrumentation, the design permits a variety of other instruments to share the common infrastructure. The general requirement on any additional scientific module is that it fits physically, does not interfere with existing equipment and is compatible with the resources available at the GLIP (for example power and communication). A number of scientific modules (for example auroral spectrometers, photometers, cloud cameras) were planned but are not yet realised. For a period, three stations, (1) Kiruna, (3) Silkkimuotka and (6) Nikkaluokta were equipped with pulsation magnetometers operated by the University of Newcastle, Australia. So far these have only used the mains power and not been remote-controlled. A radio-experiment involving a new type of antenna and a 3D-receiver is planned to be installed for testing at some of the ALIS stations [Puccio, 2002].

Scientific considerations

While large-scale auroral phenomena are most conveniently studied from space, medium to small-scale phenomena are usually studied using ground-based instruments. ALIS was designed to make absolute measurements of auroral phenomena within the field-of-view of a traditional all-sky camera (Figure 2.1).

In order to make accurate absolute measurements of auroral emissions, the optimal situation exists when the observation is carried out close to the magnetic zenith of the observation site. As the zenith angle increases, the need for photometric corrections due to spatial smearing as well as atmospheric effects also increases rapidly. On the other hand, if the field-of-view is too small, many more stations are required in order to obtain an acceptable spatial coverage and overlapping fields-of-view suitable for triangulation at auroral altitudes. Selecting a field-of-view in the range of $50^{\circ}$ to $90^{\circ}$ with a station baseline of 50-100 km appeared as a suitable compromise, and also economically feasible. Figure 2.3 illustrates

Figure 2.3: At a 50 km baseline (left) stations looking into zenith with a field-of-view of $90^{\circ}$ (green) have overlapping fields-of-view from about 25 km. Limiting the field-of-view to about $50^{\circ}$ (yellow) raises the height of overlap to 50 km. Increasing the station baseline to 100 km (right), the fields-of-view overlaps from about 50 km at $90^{\circ}$, and from about 100 km at about $50^{\circ}$ field-of-view. It is furthermore seen that it is desirable to have steerable cameras in order to image a common volume with as many stations as possible (see also Figure 3.11).

the effects of station baselines and fields-of-view in these ranges. Note that the fields-of-view under consideration in this section are along the x/y directions on the CCD, not to be confused with the diagonal, or optical field-of-view, refer to Figure 4.1 and Table 4.6 for details. It is immediately seen that a $50^{\circ}$ field-of-view combined with a 100 km baseline will not provide sufficient overlap for auroral triangulation and tomography (assuming the lower edge of the auroral curtain at about 105 km [Størmer, 1955]). Consequently, a baseline of about $50$ km was selected together with a field-of-view of about $50^{\circ}$ (see Table 4.6 for details). However, as four stations were put into operation, it was realised that increasing the field-of-view to about $60^{\circ}$ would reduce the artifacts during auroral tomography (see references in Section 6.3), as well as providing better triangulation possibilities for studies of lower lying objects, for example polar-stratospheric clouds. Locating the two $60^{\circ}$ imagers at appropriate stations thus provided a possibility for enhancing the results of tomography and triangulation.

The next parameter to consider is the spatial coverage and achievable field-of-view per pixel, $\mathit{FoV_{p}}$. Table 2.2 lists the linear coverage at some altitudes of

Table 2.2: Examples of approximative imager coverages in km (boldface), at some altitudes of interest given either a $50^{\circ}$ or $60^{\circ}$ imager field-of-view. For each field-of-view, the corresponding linear field-of-view per pixel ( $\mathit{FoV_{p}}$) and pixel-coverage (in km) for a pixel looking in the zenith direction are given. (see also Section 4.3). The number of pixels also reflects some common binning factors in use with the present six ALIS imagers (see also Figure 3.4).

			coverage in [km] at altitudes [km]:
$\mathit{FoV_{}}$	pixels	$\mathit{FoV_{p}}$	40	80	105	250	500	1000
$\mathbf{50^{\circ} }$			37	75	98	233	466	933
	64	$0.78^{\circ}$	0.55	1.09	1.43	3.41	6.82	13.64
	128	$0.39^{\circ}$	0.27	0.55	0.72	1.70	3.41	6.82
	256	$0.20^{\circ}$	0.14	0.27	0.36	0.85	1.70	3.41
	512	$0.10^{\circ}$	0.07	0.14	0.18	0.43	0.85	1.70
	1024	$0.05^{\circ}$	0.03	0.07	0.09	0.21	0.43	0.85
$\mathbf{60^{\circ} }$			46	92	121	289	577	1155
	64	$0.94^{\circ}$	0.65	1.31	1.72	4.09	8.18	16.36
	128	$0.47^{\circ}$	0.33	0.65	0.86	2.05	4.09	8.18
	256	$0.23^{\circ}$	0.16	0.33	0.43	1.02	2.05	4.09
	512	$0.12^{\circ}$	0.08	0.16	0.21	0.51	1.02	2.05
	1024	$0.06^{\circ}$	0.04	0.08	0.11	0.26	0.51	1.02

interest for both the whole field-of-view, as well as for a pixel looking in the zenith direction. Note that these values are only to be interpreted as a first order approximation (see also Table 4.6 in Section 4.3). At 105 km altitude and 1024 pixels, the achievable pixel field-of-view is in the order of 100 m in zenith.

Selecting sites for the ALIS stations

Selecting the actual sites for the stations involved compromises. Although the first paper on ALIS [Steen, 1989] assumed that some stations would have to generate their own power and rely on microwave or satellite communications, budgetary considerations required the stations to be located in the vicinity of existing power and telecommunication lines. It was decided that the first station should be located close to the Swedish Institute of Space Physics (IRF) in Kiruna, in order to simplify development. The final decision on where to locate the remaining ALIS stations was based on a careful evaluation of a number of sites with regard to station separation (about 50 km) and geometry of ALIS with respect to tomographic as well as general auroral observation requirements, the proximity to commercial electrical power, telecommunication infrastructure and road access. Another important criteria was to find sites with low levels of man-made light pollution and a reasonably free horizon.

The highest priority was to populate the Tromsø-Kiruna meridian with stations, thereafter expansions towards east and west were desired. Practical considerations led to the stations being deployed in the following order (see Figure 2.4): (1) Kiruna, (2) Merasjärvi, (3) Silkkimuotka, (4) Tjautjas, (5) Abisko and (6) Nikkaluokta. After that, an expansion southward was planned with stations (7) Kilvo and (8) Nytorp. This was mainly in order to accommodate measurements of southward expansion of the auroral oval during the upcoming solar maxima. Later the plans were changed in favour of one station in Norway, (9) Frihetsli, to be possibly followed by a station at the EISCAT site at Ramfjordmoen, Norway. The motivation for this change of plans was to give a better support to combined measurements with EISCAT. Awaiting this expansion northward, a tenth mobile station provided zenith coverage along the Tromsø magnetic field-line during active experiments with HF pump-enhanced airglow and the EISCAT radar facility (Section 6.4). A summary of site numbers, names, acronyms and geographic coordinates is found in Table 2.3 and in Figure 2.4.

Table 2.3: Geographical coordinates of the ALIS stations. Notes: Station No. 1 moved in the summer of 1999, see text. Stations No. 7-8 were deployed on site but never used. Station No. 9 was never deployed. Station No. 10 is mobile.

				latitude			longitude			h
No.	Adr.	Site name	Acronym	$^{\circ}$	$'$	$''N$	$^{\circ}$	$'$	$''E$	m
1	S01	IRF	KRN	$67$	$50$	$26.6$	$20$	$24$	$40.0$	$425$
1	S01	Knutstorp	KRN	$67$	$51$	$20.7$	$20$	$25$	$12.4$	$418$
2	S02	Merasjärvi	MER	$67$	$32$	$50.7$	$21$	$55$	$12.3$	$300$
3	S03	Silkkimuotka	SIL	$68$	$1$	$47.0$	$21$	$41$	$13.4$	$385$
4	S04	Tjautjas	TJA	$67$	$19$	$57.8$	$20$	$45$	$2.9$	$474$
5	S05	Abisko	ABK	$68$	$21$	$20.0$	$18$	$49$	$10.5$	$360$
6	S06	Nikkaluokta	NIL	$67$	$51$	$6.7$	$19$	$0$	$12.4$	$495$
7	S07	Kilvo	KIL	$$	$$	$$	$$	$$	$$	$$
8	S08	Nytorp	NYT	$$	$$	$$	$$	$$	$$	$$
9	S09	Frihetsli	FRI	$$	$$	$$	$$	$$	$$	$$
10	S10	Mobile	BUS	$$	$$	$$	$$	$$	$$	$$

Figure 2.4: Map of northern Scandinavia displaying the final locations of the ALIS stations. See also table Table 2.3. The Control-Centre as well as a secondary Operations Centre is located in Kiruna.

Station (1) Kiruna was initially located in the optical laboratory at IRF, Kiruna but had to be moved a couple of kilometres (to Knutstorp, close to the Kiruna EISCAT-site) in the fall of 1999 due to ongoing construction work and rising levels of man-made light pollution at the original site.