A two-probe method for measuring water content of thin forest floor litter layers using time domain reflectometry

Abstract

Few methods exist that allow non-destructive in situ measurement of the water content of forest floor litter layers (Oa, Oe, and Oi horizons). Continuous non-destructive measurement is needed in studies of ecosystem processes because of the relationship between physical structure of the litter and the biological and chemical processes that take place therein. We developed a method using time domain reflectometry (TDR) to monitor water content in a coniferous forest floor litter layer. Litter and mineral soil horizons were reconstructed in test beds in which TDR probes were placed and measurements taken using a range of litter and mineral soil water contents. Two probes are necessary when litter thickness is less than the spatial sensitivity (6 to 8 cm) of the TDR probes; one probe placed in the mineral soil and another one at the interface of the litter and mineral soil. Using this arrangement of TDR probes and simple mathematical relationships, the volumetric water content of forest litter can be estimated continuously. When the results of the two-probe method are compared to volumetric water content of forest litter obtained by gravimetric means there is a strong positive linear relationship between the two measured values of litter water content ( r 2 = 0.93). The two-probe method, however, underestimates litter water at low water contents and overestimates it at high water contents. This error has at least three components: (1) TDR instrument error, (2) errors in estimating volumetric water content from gravimetric data, and (3) using a TDR calibration curve not specific for high organic matter litter layer material. Calibrating the instrument for this specific condition should be improve the overall estimate of the litter layer water content.