Farm planning generally focuses on optimal diversification with respect to risk and uncertainties, where the risk-management strategies combine production, marketing, financial and environmental responses of the production of farm firm. In this study an empirical examination of farm diversification has been carried out from a sample of farms in Eastern Norway in which four measures of diversification (indices) were defined to incorporate the risk and uncertainties in relation to farm production (total) income. Using these four alternative measures of diversification and panel-data techniques, it has been shown that larger farms are more diversified, and when there is productive location and access to labour the farmers have a greater incentive to spread risk. These results suggest that diversification and farm size are positively linked and that there may not be sufficient economies of scale to warrant specialisation. Alternatively, farm specialisation may not be environmentally desirable because of the soil and water pollution that have been caused by large scale farming systems usually practised in most of the developed countries like Norway. Hence, farm diversification can be used as a strategy for managing the pollution caused by large scale less diversified farming systems, as well as to spread the risk related to farm income.