Effect of using time-lag between maximum screen temperature and soil temperature in improving annual soil regression equations

Soil temperature is an important physical parameter that affects energy balance between the atmosphere and earth’s surface. Soil and air temperature which are controlled by surface energy balance, have strong physical relationship. Because of the difference in thermal capacity between air and soil temperature, lag time exists between the occurrence of maximum of soil (Ts) and air (Ta) temperatures. In this research, observed daily soil temperature (6 depths) and screen temperature (170 cm) data were used from 1993 to 2008 for different climates including: Ahvaz (arid), Tabriz (semi-arid), Ilam (Mediterranean), Babolsar (humid) and Rasht (hyperhumid). Since, there are limited studies which have investigated the lag-time between the occurrence of maximum soil temperature (Ts) and screen temperature (Ta), this study aims to investigate the role of the lag-time on annual Ts-Ta regression relations. We constructed various regression equations between maxima of Ts (at 50 and 100 m) and Ta (170 cm). Using correlation method and Fourier series approach, the time lag between Ts (independent variable) and Ta (dependent variable) was also determined for all the study sites. Results from the mentioned methods were compared by RMSE, MAE, MBE and MPE criteria against each other. The differences between the two methods were statistically insignificant. The numbers of lag days (NLD) were found to be nearly the same for either mentioned methods. Regression equations between two statistical pairs (Ts-Ta) were constructed for two different scenarios (with and without lag) and the corresponding coefficients of determination were compared for the above scenarios. Results revealed that inclusion of NLD could significantly improve the Ts-Ta regression equations (at P<0.1 significant level), in particular for 100 cm depth. In addition, the NLD and the relevant equations were found to be different for each climate.