Incorporating Influential Factors in Diurnal Temperature Estimation with Sparse Data

In order to achieve certain research objectives, researches may have to utilize data at different levels. Temperature is one such attribute that is being collected at different levels. In many of the meteorological stations, daily temperature readings are being collected but not the hourly values. Due to the importance of hourly temperature values in many studies, many methods can be found in the literature to estimate hourly temperature reading using sparse data. Moreover, apart from daily maximum and minimum temperatures, many auxiliary variables such as sun set time, sun rise time and solar radiation values can be considered important to increase the accuracy of the estimates. This paper suggests an algorithm to incorporate influential variables when estimating hourly temperature values using sparse data. The paper also proposes a novel method "RATE" to estimate unusual temperature curves during rainy days, that have shown equal or better results when compared to LEA. However, for situations, where the rain times of the day becomes random, the RATE seems to be less accurate.