|
Ever since NORSAR was established in 1968, the focus of research has been upon developing advanced array processing techniques. Originally, NORSAR was constructed as a large aperture array, with an initial diameter of 100 km, and 22 subarrays comprising a total of 132 short period and 22 three-component long-period seismometers. NORSAR is still the largest stand-alone array in the world, even after it was reduced in size to 7 subarrays (diameter 60 km) in 1976. From the beginning, the research at NORSAR focused on processing techniques for large, so-called teleseismic arrays.
In the 1980s, emphasis shifted to smaller arrays, since the prospects of in-country stations for CTBT monitoring were emerging. This resulted in the first regional array, NORES, which was constructed in southern Norway in 1984, and had 25 seismometer sites within an area of only 3 km in diameter. A sister array, ARCES, was established in northern Norway in 1987. NORES and ARCES have since become the standard for arrays in the International Monitoring System.
Seismic arrays and their relative size.
The relative size of the arrays is illustrated in the figure above. Naturally, different processing techniques are required for arrays of such vastly different sizes. In the case of the large NORSAR array, the signals are only partially coherent across the array, and therefore standard plane-wave delay and sum beamforming does not provide optimum processing. On the other hand, the smaller arrays often exhibit coherent noise, and this must be appropriately taken into account when devising processing methodology.
Ïn the articles below, we will first briefly address the small-array processing techniques, which are still the main area of focus. However, we will also describe some important new developments at NORSAR relating to large-array processing.
|
