An Automatic Processing Framework for In Situ Determination of Ecohydrological Root Water Content by Ground-Penetrating Radar

Abstract

Root water content (RWC) is a vital component in water flux in soil–plant–atmosphere continuum. Knowledge of RWC helps to better understand the root function and the soil–root interaction and improves water cycle modeling. However, due to the lack of appropriate methods, field monitoring of RWC is seriously constrained. In this study, we used ground-penetrating radar (GPR), a common geophysical technique, to characterize RWC of coarse roots noninvasively. An automatic GPR data processing framework was proposed to (1) identify hyperbolic root reflections and locate roots in GPR images and (2) extract waveform parameters from the reflected wave of identified roots. These waveform parameters were then used to establish an empirical model and a semiempirical model to determine RWC. We validated the developed models using GPR root data at three antenna center frequencies (500 MHz, 900 MHz, and 2 GHz) that were produced from simulation experiments (with RWC ranging from 70% to 150%) and field experiments in sandy soils (with RWC ranging from 66% to 144%). Our results show that both the empirical and the semiempirical models achieved a good performance in estimating RWC with similar accuracy, i.e., the prediction error [rootmean-square error (RMSE)] was less than 8% for the simulation data and 12% for the field data. For both models, the accuracy of RWC estimation was the highest when applied to 2-GHz data. This study renders a new opportunity to determine RWC under Manuscript received January 3, 2021; revised February 9, 2021; accepted March 5, 2021. This work was supported in part by the National Natural Science Foundation of China under Grant 41401378 and Grant 41571404, and in part by the “Fundamental Research Funds for the Central Universities” at Sichuan University under Grant YJ202087 and Grant YJ202093. The work of Dedi Yang was supported by the United States Department of Energy to Brookhaven National Laboratory under Contract DE-SC0012704. (Corresponding author: Bihang Fan.) Xinbo Liu, Xihong Cui, and Jin Chen are with the Faculty of Geographical Science, Beijing Normal University, Beijing 100875, China (e-mail: [email protected]; [email protected]; [email protected]). Li Guo and Bihang Fan are with the State Key Laboratory of Hydraulics and Mountain River Engineering, College of Water Resource and Hydropower, Sichuan University, Chengdu 610065, China (e-mail: [email protected]; [email protected]). John R. Butnor is with USDA Forest Service, Southern Research Station, 81 Carrigan Drive, Aiken Center, University of Vermont, Burlington, VT 05405 USA (e-mail: [email protected]). Elizabeth W. Boyer is with the Department of Ecosystem Science and Management, Penn State University, University Park, PA 16801 USA (e-mail: [email protected]). Dedi Yang is with the Department of Ecology and Evolution, Stony Brook University, Stony Brook, NY 11794 USA, and also with the Environmental and Climate Sciences Department, Brookhaven National Laboratory, Upton, NY 11973 USA (e-mail: [email protected]). Color versions of one or more figures in this article are available at https://doi.org/10.1109/TGRS.2021.3065066. Digital Object Identifier 10.1109/TGRS.2021.3065066 field conditions that enhances the application of GPR for root study and the understanding and modeling of ecohydrology in the rhizosphere.