第四章 英汉论文摘要翻译.docVIP

Discussion and conclusions The main purpose of this study was to develop a new method to estimate root biomass using root diameter and average root density. According to the results of the field experiment, for coarse root of the same tree species with diameters larger than 0.5cm, estimated root biomass had good agreement with measured values (R20.8). Results suggest that the new method can feasibly estimate root bi-omass from root diameter estimates by GPR if the root den-sity is approximately consistent. The key factor in estimating biomass is the precise esti-mation of root diameter by GPR. Thus, we developed a root diameter estimation model based on parameter ΔT from 2 GHz frequency antennas. The major advantage of parameter ΔT is that it only depends on the geometric dimension of the reflector and is independent of signal strength. As a result, in the radar profiles, ΔT is only related to the shape of the reflected wave and is independent of amplitude, which de-creases with root depth. Accordingly, it could be used to precisely estimate root diameter without considering the effects of radar wave attenuation. For the same reason, root biomass estimation based on estimated diameter by the GPR model is also unaffected by root depth or radar signal at-tenuation. This resulted in high consistency between esti-mated and measured values with an R2 larger than 0.8. Batnor et al. [20] used signal intensity to estimate root di-ameter and biomass, in which the determination coefficients between the extracted parameter and root diameter at depths of 15cm and 30cm were 0.81 and 0.55, respectively, and the determination coefficient with root biomass was only 0.55. This was primarily due to the influence of radar signal energy attenuation with depth [20]. In subsequent studies, researchers improved the radar data process method and obtained a determination coefficient of 0.68 between the parameter and root biomass; however, it only applied to roots at the same depth [22–24].