Sensitivity of Jw to errors in model inputs of heat pulse method
Description
Description: The inaccuracies in C and λ estimates and probe spacing measurements (i.e., xd and xu) may lead to unknown errors in thermo-TDR measured Jw values derived from the Kluitenberg et al. (2007) model. Therefore, we conducted a theoretical error analysis on a hypothetical soil to quantify the sensitivity of Jw to errors in these model inputs.
Files
Steps to reproduce
The inaccuracies in C and λ estimates and probe spacing measurements (i.e., xd and xu) may lead to unknown errors in thermo-TDR measured Jw values derived from the Kluitenberg et al. (2007) model. Therefore, we conducted a theoretical error analysis on a hypothetical soil to quantify the sensitivity of Jw to errors in these model inputs. The hypothetical soil was assumed to be a loam with 40% sand, 40% silt, and 20% clay. The initial situations for ρb and θw of the soil were set as 1.0 Mg m-3 and 0.3 m3 m-3, respectively. We considered the hypothetical soil under 10 different transient water flow conditions with Jw ranging from 3.6 to 36000 mm h-1. Using the assigned model parameters and Jw values, the temperature changes at the upstream and downstream needles were estimated inversely with the Kluitenberg et al. (2007) model. We then applied random errors (all set at +5% of the initial values) to model inputs of C, λ, and xd separately. The relative deviations from the initial Jw were used as a criterion to assess the sensitivity of thermo-TDR method to errors in the model inputs.