Towards a general equation for frequency domain reflectometers

26 It is well documented that capacitance-based soil moisture sensor measurements are particularly influ27 enced by particle size distribution, density, salinity, and temperature of a soil, in addition to its moisture 28 content. Moreover, the equations provided by manufacturers of soil moisture sensors are often only 29 applicable to a limited number of soil types, thus yielding significant errors when compared with gravi30 metric measurements for observations in real soils. This limitation makes site-specific calibrations of 31 such sensors necessary. Consequently, development of a general equation provides the possibility to 32 derive the needed parameters from information such as soil type or particle size distribution. This paper 33 describes the development of a general equation for the Campbell Scientific CS616 Water Content Reflec34 tometers using data from sensors installed throughout the Goulburn River experimental catchment. It is 35 subsequently tested using monitoring sites in the Murrumbidgee Soil Moisture Monitoring Network, 36 which were not part of the original development; both monitoring networks are located in south-eastern 37 Australia. Previously developed equations for temperature correction and soil moisture estimation using 38 the Campbell Scientific CS615 Water Content Reflectometer are adapted to the new CS616 sensor. More39 over, relationships between readily available soil properties and the parameters of the general equations 40 are derived. It is shown that the general equations developed here can be applied to data collected in the 41 field using only information on the soil particle size distribution with an RMSE of around 6% m/m 42 (<1% m/m under laboratory conditions; which is a significant improvement in comparison to 14% 43 m/m when using the manufacturer’s equations). 44 2010 Published by Elsevier B.V.