Haruyuki Fujimaki

Assessment of some soil thermal conductivity models via variations in temperature and bulk density at low moisture range

Simulation of heat transfer in soil under steady and unsteady situations requires reliable estimate of soil thermal conductivity (λ) at varying environmental conditions. In the current work several soil thermal conductivity predicting models including I) de Vries, II) Campbell, III) combined de Vries and Campbell and IV) de Vries-Nobre were evaluated for the four soils of coarse sand, sandy loam, loam and clay loam textured at varying in temperature and bulk density at low moisture range. Thermal conductivities measured by the cylindrical probe method served as the reference for models assessment. Results showed that approximately same thermal conductivities obtained by the five methods at low moisture range (θ ≤ 0.05 m3/m3). Also the de Vries and de Vries-Campbell models produced accurate than Campbell and de vries-Nobre models. The accuracy of the two models increased with soil compaction but decreased with temperature rise. Campbell model showed more reliability at higher (311.16 and 321.16 K) temperatures; but its accuracy declined with soil compaction in current work. It seems that assuming needle shape for the soil particles is far away from the reality whereas assuming spherical shapes may be more realistic and produced more satisfactory prediction of thermal conductivity. The compaction would alter particle arrangement and may increase the contact area of particles; and then make them behave more or less spherical shape.it seems thermal conductivity in solid particles increase via increasing in temperature. Since a modified mineral shape factor, gm, was developed as a combination between sphere and needle according to geometric mean particle diameter as well as bulk density and temperature as modifying factors. This factor increased the accuracy of de Vries-Nobre model up to 10.37%. Regarding nonlinear regression model, moisture content, bulk density, temperature and quartz content demonstrated significant effect on soil thermal conductivity in our investigation.

Tracking Infiltration Front Depth Using Time-lapse Multi-offset Gathers Collected with Array Antenna Ground Penetrating Radar

A Ground Penetrating Radar (GPR) system based on a ground-coupled, densely populated antenna array was used to collect data during an infiltration experiment conducted at a test site near the Tottori Sand Dune, Japan. The antenna array used in this study consists of 10 transmitting antennas (Tx) and 11 receiving antennas (Rx). For this experiment, the system was configured to use all possible Tx-Rx pairings, resulting in a Multi-Offset Gather (MOG) consisting of 110 Tx-Rx combinations. The array was left stationary at a position directly above the infiltration area and data were collected every 1.5 seconds using a time-based trigger. Common-Offset Gather (COG) and Common Mid-Point (CMP) data cubes were reconstructed from the MOG data during post-processing. There have been few studies that used time-lapse CMP data to estimate changes in velocity of propagation. In this study, electromagnetic (EM) wave velocity was estimated heuristically at 1-minute intervals from the reconstructed CMP data through curve fitting, using the hyperbola equation. We then proceeded to calculate the depth of the wetting front. The evolution of the wetting front over time obtain through this method is consistent with the observations from a soil moisture sensor which was placed at a depth below 20 cm. The results obtained in this study demonstrate the ability of such array GPR system to monitor a subsurface dynamic process like water infiltration accurately and quantitatively.

A new scheme to optimize irrigation depth using a numerical model of crop response to irrigation and quantitative weather forecasts

Abstract Irrigation management can be improved by utilizing advances in numerical models of water flow in soils that can consider future rainfall by utilizing data from weather forecasts. Toward this end, we developed a numerical scheme to determine optimal irrigation depth on scheduled irrigation days based on a concept of virtual net income as a function of cumulative transpiration over each irrigation interval; this scheme combines a numerical model of crop response to irrigation and quantitative weather forecasts. To evaluate benefits, we compared crop growth and net income of this proposed scheme to those of an automated irrigation method using soil water sensors. Sweet potato (Ipomoea batatas (L.), cv. Kintoki) was grown in 2016 in a sandy field of the Arid Land Research Center, Tottori University, Japan under either a non-optimized automated irrigation or the proposed scheme. Under the proposed scheme, 18% less water was applied, yield increased by 19%, and net income was increased by 25% compared with the results of the automated irrigation system. In addition, soil water content simulated by the proposed scheme was in fair agreement with observed values. Thus, it was shown that the proposed scheme may enhance net income and be a viable alternative for determining irrigation depths.