Clouds play an important role in the climate system, and strongly modify radiation conditions in the atmosphere and at the earth's surface. Present estimations show that the net impact of clouds in the atmosphere results in a cooling several times larger than the expected warming that would result from a doubling of the CO2-concentration in the atmosphere. Regional and global monitoring of cloud conditions is therefore necessary for studying the role of clouds in possible climate feedback mechanisms. This paper presents a method to estimate cloud conditions (cloud cover) in the Nordic region, from multispectral satellite data. A cloud classification scheme, based on high-resolution imagery data from polar orbiting NOAA-satellites, was used to produce monthly cloud frequencies for the summer of 1993. Comparisons with existing surface observations have shown very good agreement. Cloud conditions are shown to be highly sensitive to characteristics of the earth's surface. Large differences between land and sea areas were found, especially in the beginning of summer. Cloud frequencies were significantly lower over the Baltic Sea compared to surrounding land areas. Mountainous areas showed, on the contrary, much higher cloud frequencies than surrounding areas.