Evaluation Of Nine Electromagnetic Soil Water Content Sensors For Water Content Determination In Sawdust And Tree Trunk

Plant transpiration is an important part of the water cycle in terrestrial ecosystems,which controls land surface energy balance,determines watershed hydrological processes,and affects regional and global climate.Plant stem is an important water transport channel in the soil plant atmosphere continuum.The water stored in the stem can not only satisfy the physiological activities of plants,but also improve the ability of plants to cope with climate stress.Therefore,accurate measurement of stem water content is not only beneficial to better understand the physiological conditions of trees and study the response mechanism of vegetation under climate change,but also beneficial to formulate appropriate irrigation management measures and improve tree productivity.The commonly used methods for measuring stem water content(e.g.,drying method,gamma ray attenuation method,nuclear magnetic resonance method,and ultrasonic detection method)are time-consuming and destructive.There is no continuous,automatic,and in-situ monitoring method to measure stem water content.At present,soil water measurement technology has been highly developed.Although soil and tree stem are both porous media,studies have shown that there may be errors in measuring stem water by direct use of electromagnetic soil water sensor.The objective of this study was to evaluate the applicability of electromagnetic pulse soil water sensors for measuring volumetric water content of tree stems.In this study,nine electromagnetic soil water sensors(i.e.,EC-5,10 HS,TEROS 10,TEROS 12,Hydra Probe,Theta Probe ML2 X,TDR310S,TDR315 L,and TLO)were selected.Their performance was evaluated for water content determination of three kinds of wood dusts(i.e.,poplar,Chinese fir,and cypress),and based on which the calibration curves were established.In addition,five short-needle electromagnetic soil water sensors(i.e.,TEROS 10,TEROS 12,Hydra Probe,Theta Probe ML2 X,and TDR310S)which can be easily inserted into tree stems were used to measure the volumetric water content of five tree stems(i.e.,poplar,willow,walnut,sophora,and Chinese toon).The main conclusions are as follows:(1)In this study,9 electromagnetic soil water sensors were used to measure the volumetric water content of 3 kinds of sawdusts with predetermined bulk densities and water contents.Each sawdust included 3 kinds of bulk densities and 8 water contents(from dryness to saturation).The results showed that the electromagnetic soil water soil moisture sensors can be used to measure the volume water content of wood chips after correction,and the calibration error of TDR310 S sensor was the smallest and the calibration effect was the best root-mean-squared-error(RMSE=0.022 cm3 cm-3).The calibration curve of TLO sensor has the smallest correction error and the best correction effect(RMSE=0.016,0.022cm3 cm-3)in the volume water content of wood chips of fir and pine,respectively.The calibration curve of TDR310 S sensor has the best fitting curve(R2=0.98)and the smallest error(RMSE=0.023 cm3 cm-3)among the three kinds of saw dusts.The original calibration curve of the sensor is not applicable to measure water content.Therefore,new calibration curve needs to be established for sawdusts.(2)Five short-needle soil water sensors were used to measure the volumetric water content of the stems of five species in this study.Each stem underwent a drying-wetting-drying process.The results showed that TEROS series sensors were less affected by tree species.The fitting curves of TEROS 12 were better(R2 >0.96),and the correction curve error of TEROS 10 was smaller(RMSE ≤ 0.028 cm3 cm-3)after correction.The performance of Hydra Probe and Theta Probe in stem water content measurement was poor(RMSE all≤0.028 cm3 cm-3).At the same time,considering the difficulty of pre-drilling,TEROS 10,TEROS 12 are ideal sensors for measuring water content of tree stem.In addition,empirical correction for different tree species is necessary for accurate measurement of volumetric water content of stem.(3)Based on bulk density,this study established a semi-empirical prediction model for sawdust and stem volume water content,and found that the correction taking bulk density into account volume water content could improve the accuracy of soil moisture sensor measurement in most cases.In the semi-empirical model considering wet bulk density and dry bulk density,the quadratic polynomial curve fitting established by Theta Probe based on toon and sawdust data had the best fitting effect(R2=0.94),and the root mean square error was(RMSE=0.044,0.059 cm3·cm-3),respectively.In addition,the R2 of semi-empirical model considering dry bulk density and wet bulk density has little difference.In summary,this study first systematically assess the applicability of various electromagnetic soil water sensors to determine water content of sawdust and stem.The calibration curve to estimate the volumetric water content are established for each sensor type and investigated tree species.Accurate determination of volumetric water content of saw dusts and trees sections provide scientific guidance and reference and technical support for the study of plant water utilization and soil-plant-atmosphere water exchange continuum.