TY - JOUR

T1 - Characterizing uncertainty in process-based hydraulic modeling, exemplified in a semiarid Inner Mongolia steppe

AU - Zhao, Ying

AU - Wang, Haixia

AU - Song, Bing

AU - Xue, Pengfei

AU - Zhang, Wangchen

AU - Peth, Stephan

AU - Lee Hill, Robert

AU - Horn, Rainer

N1 - Funding Information: This work was supported by the Natural Science Foundation of China (41977009), the Taishan Scholars Program, China (201812096), the Cooperation Project of Key Laboratory of Chinese Academy of Sciences, and the German Research Council (DFG) in the framework of the Interdisciplinary Research Project MAGIM (Matter fluxes in grasslands of Inner Mongolia as influenced by stocking rate), and the Key Program of the National Natural Science Foundation of China (42320104006).

PY - 2023/12

Y1 - 2023/12

N2 - Assessing root sources of three uncertainties – parameterization of soil hydraulic characteristics, boundary conditions, and estimation of source/sink terms – is a significant challenge in soil water transport modeling. This study aims to evaluate the uncertainty of three each widely-used parameter estimation methods affecting plot-scale water dynamics. The study employs HYDRUS, a process-based hydrologic model, to incorporate these uncertainties and compare model predictions to measured values in a semiarid Inner Mongolia steppe, China. Soil hydraulic parameters are determined using two direct methods (laboratory-derived approach and evaporation method) and one indirect method (neural network). While each hydraulic parameter method generally simulates soil moisture dynamics, the evaporation method performed better, especially under dry conditions. This suggests that measuring the intensity properties, such as unsaturated hydraulic conductivity, with the evaporation method is crucial for reasonable soil moisture simulation. The study also demonstrates the impact of different applied boundary conditions on simulated soil moisture, specifically the partitioning of reference FAO evapotranspiration via one direct method (soil fraction cover) and two indirect methods (leaf area index and crop height). The partitioning via soil fraction cover reflected a better simulation. Additionally, the study compares the uncertainties of root water uptake function with root growth parameters and constant root depth referenced to grass and pasture, and finds no significant difference among them. Comparing three sources of uncertainty in predicting soil moisture, the study concludes that the input soil hydraulic parameter is more sensitive than evapotranspiration partitioning or representation of root water uptake function. Our study highlights that measuring soil intensity properties can better reflect the effects of land use change, such as compaction, on field water transports.

AB - Assessing root sources of three uncertainties – parameterization of soil hydraulic characteristics, boundary conditions, and estimation of source/sink terms – is a significant challenge in soil water transport modeling. This study aims to evaluate the uncertainty of three each widely-used parameter estimation methods affecting plot-scale water dynamics. The study employs HYDRUS, a process-based hydrologic model, to incorporate these uncertainties and compare model predictions to measured values in a semiarid Inner Mongolia steppe, China. Soil hydraulic parameters are determined using two direct methods (laboratory-derived approach and evaporation method) and one indirect method (neural network). While each hydraulic parameter method generally simulates soil moisture dynamics, the evaporation method performed better, especially under dry conditions. This suggests that measuring the intensity properties, such as unsaturated hydraulic conductivity, with the evaporation method is crucial for reasonable soil moisture simulation. The study also demonstrates the impact of different applied boundary conditions on simulated soil moisture, specifically the partitioning of reference FAO evapotranspiration via one direct method (soil fraction cover) and two indirect methods (leaf area index and crop height). The partitioning via soil fraction cover reflected a better simulation. Additionally, the study compares the uncertainties of root water uptake function with root growth parameters and constant root depth referenced to grass and pasture, and finds no significant difference among them. Comparing three sources of uncertainty in predicting soil moisture, the study concludes that the input soil hydraulic parameter is more sensitive than evapotranspiration partitioning or representation of root water uptake function. Our study highlights that measuring soil intensity properties can better reflect the effects of land use change, such as compaction, on field water transports.

KW - Evapotranspiration

KW - Root water uptake

KW - Soil moisture simulation

KW - Uncertainty analysis

KW - Unsaturated hydraulic conductivity

UR - http://www.scopus.com/inward/record.url?scp=85177741806&partnerID=8YFLogxK

U2 - 10.1016/j.geoderma.2023.116713

DO - 10.1016/j.geoderma.2023.116713

M3 - Article

AN - SCOPUS:85177741806

VL - 440

JO - GEODERMA

JF - GEODERMA

SN - 0016-7061

M1 - 116713

ER -