PREDICTING SOIL MOISTURE AND SOIL TEMPERATURE IN A TROPICAL PEATLAND USING WATER TABLE DEPTH, SURFACE TEMPERATURE AND RAINFALL

Abstract

Degradation of tropical peatland largely depends on water table depth and other soil physical properties like soil moisture and soil temperature. Empirical studies indicate positive relationships between water table depth, soil moisture and soil temperature. This paper aims to further investigate the relationship between soil moisture and soil temperature variability in time as a function of soil depth and their relationship with water table depth. Linear models for prediction of soil moisture and soil temperature at 70 cm using surface soil moisture and soil temperature measurements, rainfall and surface temperature are examined. In all the four plots used in the study, seasonality of soil moisture and soil temperature was confirmed as they strongly correlate with rainfall. Both soil moisture and soil temperature at 5 cm, as a result of storm events, experienced more fluctuations compared to the values recorded at 70 cm. The linear models give better prediction capabilities as predicted and observed values give good fits and coefficients of determination, R2, is mostly above 0.5 with agreements within 0.05 – 0.1.