Variability of soil thermal diffusivity in three Iowa fields

Abstract

AbstractSoil thermal properties influence the rates of chemical and biological reactions in the soil, seed germination, crop establishment, and productivity. Soil temperature varies with time and depth, and thermal diffusivity (α) is the main property associated with soil temperature variations. Variations in α within row‐cropped fields under long‐term tillage practices are not well documented in Iowa fields. The objective of this study is to determine α variations within three central Iowa fields under long‐term tillage practices and to estimate the required number of measurement sites within each field to determine α at a specified precision. Each field receives different tillage operations; one field is fall moldboard plowed followed by spring disking (MP), another field is fall chisel plowed followed by spring disking (CP), and the third field is ridge no‐tilled (RN) slot planted. Thermocouples are used to measure soil temperature at two depths, that is, 1 and 10 cm, at 49 locations (49 grid locations; 1.5 m × 3 m) in each of three adjacent fields. After allowing the thermocouples to equilibrate with the soil environment for 2 weeks in each field, hourly 1‐ and 10‐cm depth soil temperature observations were recorded for a separate 24‐h period in each field following rainfall events. Soil temperature measurements with the amplitude and phase equations were analyzed, and 49 α values were determined in each field. The larger the bulk density, the larger the thermal diffusivity. MP had the largest α values, while CP had the lowest values. The mean and standard deviation values for α in the MP, CP, and RN fields were 23 ± 4.8, 13 ± 2.2, and 21 ± 4.0 cm2 h−1, respectively. The coefficients of variation were similar for the three fields, ranging from 17% to 21%. The minimum number of samples needed to obtain a confidence interval of 7 cm2 h−1 for α was two, five, and seven, respectively, for the CP, RN, and MP fields. The MP field had the largest mean α value and required the largest number of samples to determine α at a specified precision.