Determination of soil moisture content using wavelet analysis, the case of Bule Hora wored Borena zone, Ethiopia

Abstract

In this study by designing SMS a moisture contents of different soil samples were determined using computer technique. The soil moisture sensor has been developed using the basic property that the resistance of the soil between two points decreases with the increase of water content. The SMS sensor designed to give sound signal and the signals produced were recorded and analyzed using continuous wavelet transform with MATLAB software

Five 50 gm Oven–dried soils samples were used for the experiment from the collected soil samples of Bule hora wored. The samples were watered in the range of 0ml to 50ml at 5ml interval. The result obtained show that as the water level added to the soil increases all parameters such as frequency, Amplitude and energy of the signal increased except the scale parameter of the wavelet coefficients. Correlation relation between soil water contents with electrical conductivity, maximum amplitude with electrical conductivity, frequency with electrical conductivity, and frequency with soil water contents for all the sampled soils were studied using regression with matlab. The relation between dependent variable and independent variable shows that a high correlation relation with R² greater than 0.97 and with 95% confidence level. In general the research tried to show how wavelets can be used to analyse the variation of soil moisture content in any soil types which appear to be more useful than using traditional method of soil moisture determination.