Abstract: Tank temperature distribution has influence on metering accuracy and consequently leads to metering errors. To solve this problem, the heat transfer process between the oil products inside the tank and the ambient environment was analyzed in this paper, and then a tank temperature drop model was established in consideration of numerous influencing factors, such as environment temperature, physical properties of oil products, natural convection of oil products inside the tank and solar radiation. Taking the 1^# floating roof tank of one reserve as an example, the tank temperature drop model was simulated by Fluent software, and the calculated temperature was compared with the measured temperature. The variation laws of the difference between the average vertical oil temperature at the gauging hatch and the overall average oil temperature inside the tank were figured out. Under the same ambient conditions, the lower the liquid level is, the greater the temperature drop rate is. The higher the liquid level is and the longer the storage lasts, the larger the difference between the average oil temperature at the gauging hatch and the overall average oil temperature is. The approximate range and variation trend of the above mentioned errors were displayed quantitatively by performing case calculation in different working conditions. It can be used as the reference for the metering and inventory of oil depots.