Sea ice melt processes during the Arctic summer affect the sea-air-interactions dramatically. The determination of puddles and quantification their morphology, size distribution and life cycle provides important information to the study of physical processes of sea ice, heat budget, total ice concentration etc.During the summer cruise of ARK IX/2 and 3 of the RV "POLARSTERN" in June and July 1993, a helicopter survey of sea ice for different times and locations was carried out using a Line Scan Camera (LSC) with a reselution of 2 m^2. An area between 78°N-82°N and 2°E-14°W was covered.Indication of the appearance of free water within the snow pack is usually observed in the surface layers first. This occurred on day 146 (26 May) ranging in the amount of free water present between 0.5 and 2.5%. The first melt puddles on first year ice (FY) were visually observed on June 13 day 164. The subsequent indication of the general onset of melt occurred on 17 June (Julian day 168) in the polynya region as observed on the SSM/I ice charts. This continues for about a week (Julian day 175) until the free water has increased to 6-10% within the surface layer. This period of significant free water production through melt processes coincides with a sharp increase in the melt puddles coverage, based on visual observations from the helicopter, going from 3% (Julian day 164) to 10-80% melt water on FY ice (Julian day 185-190) and 30-40% melt puddles on old ice for the same period. About 10-15 days after the brightness temperature reaches nearly black body behaviour, the maximum melt puddles coverage is reached. The refreezing period started at the end of July. The minimum of puddles concentration was detected between 80°N and 81°N. No other geographical correlation was found.Size of the puddles was between 2 m2 to more than 1 km2 and the depth was between a few centimeters to full thickness of the ice floe (thaw hole). The maximum frequency of the puddles size distribution was by a 1000 m2 category at the beginning, then shifted to 1 km2 in the middle, and finally to 100 m2 at the end of the campaign.The average depth of the snow at the beginning of ARK IX/2 varied significantly over old ice (OI) as compared to first-year ice. The mean depth during the first 10 days (Julian day 144 - 154) was 38.1 - 21.8 cm over OI and 14.6 - 8.1 cm over FY ice. The maximum-minimum range over OI varied between 62-8 cm and 26-6 cm for FY ice. By Julian day 180 the snow over FY ice had disappeared, while the snow over old ice reached that state around day 188.The comparison between SSMI and LSC ice date shows, that SSMI derived total ice concentration was higher than that from LSC but lower than the sum of both fractional concentrations of puddles and ice derived from LSC. This is what should be expected and means, that the SSMI derived ice concentrations should be corrected due to the melt features.Using the LSC provides useful ice data on small scale (resolution about 2 m2 ), which could be used for better understanding the physical processes of the sea ice.