Since a total of 73 telescopes, from small- to large-scale, participated in the ground observation, I cannot present all their results in detail here. I would like to focus on the results obtained from three large telescopes at the summit of Mauna Kea in Hawaii, of which large-aperture telescopes were able to observe the comet just after the impact. For example, the Keck Telescope, which has a 10m effective diameter using 36 split mirrors, made fluorescence-line observation of gas molecules at near infrared in an attempt to identify the composition and temperature of ice constituents, which evaporated on impact. The Subaru and Gemini telescopes, which have 8 m effective diameters with a single mirror, made observations at mid-infrared wavelengths, which are not covered by the instruments on the spacecraft. The purpose of the observation of the two telescopes was to look at the comet’s rock and nonvolatile-carbon constituents (hereafter, both collectively referred to as “dust”), which did not evaporate on impact, to measure their mineralogical composition, the ratio of crystallization, and particle size distribution. They were also expected to contribute to the understanding of the formation mechanism of the crater based on the total amount and spatial distribution of the dust. Because the Gemini and our Subaru telescope groups had independently planned almost identical observations, the two groups discussed to share observations on the event in order to optimize the overall scientific output. Since the Subaru telescope is superior in spatial resolution, though the two have the same aperture, it took the role of imaging observation while the Gemini took on spectroscopic observation.
Observations of the collision provided important discoveries. First, silicate dust illuminant at around 10μm} began to shine strongly just after impact, which had not been observed at all before the impact (Fig. 2). This shows that a large amount of fine silicate particles (from 1μm to submicron in diameter) was excavated from the interior of the comet. Second, from observations for absolute value and temporal change of emission intensity at this 10μm band, it became clear that approx. 106kg dust was released into the outer space. It was also confirmed that the dust release took place basically only at the moment of impact and that a sustained dust release over a long period of time was not induced by the event. Calculated from the observation, the total released amount of dust obtained from our observation was near the maximum value of the pre-impact predictions, suggesting that the strength of the comet's surface materials is very low. Based on the value of 106kg, it was also estimated that the diameter of the crater must be about 100m. This means that the materials excavated by the impact were from several to 10m of depth in the comet.
From spectroscopic analysis for the 10μm band, it was revealed that the silicate particles inside the comet have a very high crystallization ratio and show power-law size distribution with an exponent approximately -3.5. These results differ greatly from the previous observations of short-period comets, one of which is Tempel 1. In contrast, they are very close to the observation of long-period comets. When we were about to report our findings, the Gemini group also reached the same conclusion independently. Since the conclusion was obtained simultaneously and independently by the Subaru and Gemini groups, it is a highly reliable result. Meanwhile, the Keck telescope group reported that post-impact emergence of organic molecular species that have been found only in long-period comets in the past. This finding also supports our observational result that the internal substances of Tempel 1, a short-period comet, are very similar to those of regular long-period comets.