Due to the changes in the climate, more extreme weather conditions occur. Weather extremes (such as heat, drought, freeze, etc.) are limiting conditions for the cultivation of agricultural crops. Long term monitoring of particular quantities such as temperature, humidity, wind conditions and so on are essential for decision making in agriculture. Since the monitoring is an expensive process, the physical monitoring stations are usually very coarse. The common approach to combine these data with a global model. This paper presents steps which can be done further, using temperature as an example. The paper first presents a workflow, how to create a denser model of temperature distribution from a course one model of temperatures. Next, the paper outlines a method of how to analyze the temperature spatial distribution in time. The aim is to use historical meteorological series (e.g. of temperatures) to help farmers and producers of agriculture’s products with decision making.