Actual Evapotranspiration (AET) is at the core of the hydrological cycle. Developing accurate estimations of AET is therefore foundational to hydrological studies and the understanding and management of the hydrological cycle. Despite the importance of AET and accurate estimation, obtaining reliable values for AET remains a significant challenge for hydrological engineers. Theoretical methods often used to estimate AET, such as potential evapotranspiration (PET), have a wide range of uncertainties. Lysimeter remains the most accurate and direct AET measurement. However it requires significant labour work and funding to establish and maintain; in addition, lysimeters only measure AET at the location of the lysimeter, and therefore the data requires modification to represent AET for the entire catchment area.
Our finding research at a small stream in Toenepi, New Zealand has demonstrated a clear difference in flow rates between day and night hours. Assuming an unchanging groundwater discharge for a daily period, the daily flow fluctuation measured should represent a difference in AET rate between day and night. While several studies have investigated diurnal-fluctuation of the water table (1–16) and soil moisture (17–20), streamflow daily-fluctuation has not received enough attention so far. To address these issues, we offer a new way to estimate a catchment average for AET that uses signal processing on the daily fluctuation of streamflow hydrograph during recession curves.