Location: Soil, Water & Air Resources ResearchTitle: Improving soil heat flux accuracy with the Philip correction technique
|TONG, BING - Nanjing University Of Information Science And Technology (NUIST)|
|Sauer, Thomas - Tom|
|GAO, ZHIQIU - Chinese Academy Of Sciences|
|XIAO, XINHUA - Alabama A & M University|
|HORTON, ROBERT - Iowa State University|
Submitted to: Journal of Hydrometeorology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/22/2019
Publication Date: 7/23/2019
Citation: Tong, B., Sauer, T.J., Gao, Z., Xiao, X., Horton, R. 2019. Improving soil heat flux accuracy with the Philip correction technique. Journal of Hydrometeorology. 20(7):1435-1448. https://doi.org/10.1175/JHM-D-18-0243.1.
Interpretive Summary: The energy in sunlight that reaches crops can either be reflected, used by photosynthesis, evaporate water, heat the air and canopy, or warm the soil. The most common way to measure the amount of energy entering the soil is to use a sensor that is buried in the soil. This sensor, called a heat flux plate, uses the temperature difference from its top to bottom to estimate how much energy is moving through the plate and the surrounding soil. A scientist named Philip pointed out that the plate has a fixed ability to transmit heat while the soil's ability to transmit heat changes with changing water content. He proposed a method to correct flux plate measurements. The objective of this study was to test this correction technique for several types of flux plates under field conditions. The results indicated that large errors can occur when using flux plates and that the Philip correction needed accurate input values for plate and soil properties and didn't always improve plate estimates. These findings are of interest to scientists and manufacturers of sensors wishing to improve estimates of energy partitioning and soil heat flow.
Technical Abstract: Soil heat flux (Gs) is an important component of the soil surface energy balance. Soil heat flux plates (SHFPs) have been used widely to measure Gs, while several errors are known to occur. Philip proposed a correction that has been applied to minimize errors in Gs measured by SHFP (Gp) if the soil thermal conductivity (Ks), SHFP conductivity (Kp) and plate geometry function (H) are known. The Gp is corrected by the ratio of Gp to the predicted G ratio (Gp/Gs), which is a function of Ks/Kp and H. However, the Philip correction does not always work well. The objective of this study was to evaluate the effectiveness of the Philip correction for a variety of SHFPs of widely contrasting designs. The Kp values of four commercially available SHFPs were calibrated in agar without thermal contact resistance in laboratory experiments at flux densities ranging from 20 to 175 W m-2. The measured Kp values ranged from 4% greater to 51% less than the manufacturer-specified values. The full and simplified H expressions with measured and manufacturer-specified specifications and their impacts on the G ratio were examined. The simplified H of square SHFPs was similar to the full H value, while the simplified H of circular SHFPs was less than the full H value, and the differences increased as dimension ratios increased. The H values ranged from 0.78 to 0.88, and the differences of H were within 19%. The G ratio was sensitive to Ks/Kp and H when they were relatively small. The Gp values measured under a full corn canopy in the field underestimated gradient measured Gs (Gs_grad) values from 32% to 55%. After the Philip correction was applied, all of the Gp values agreed better with the reference Gs values. Generally, the Gp values corrected with measured plate characteristics and the full H agreed better with Gs_grad than those corrected with simplified H. The Gp values corrected with simplified H value were similar to the full H results for some SHFPs but differed for some SHFPs. These results indicated that SHFPs always underestimated Gs, and the performance of the Philip correction was affected by Kp, plate dimensions and H expression, and accurate calibration and use of the full H expression are needed.