Long-Term Simulations of the Hydrology for Sugarcane Fields in the Humid Tropics: Case Study on Guyana's Coastland

A 4.2-ha sugarcane field in Guyana was instrumented to measure hydrometeorological variables and water table depths in order to calibrate DRAINMOD. The model performs favorably, with a Nash-Sutcliffe model efficiency (NSE) of 0.72, index of agreement (IoA) of 0.92, mean absolute error (MAE) of 16.5 cm and percentage bias (PBIAS) of 1.0%. DRAINMOD is then used for long-term simulations of the field hydrology using historical climate data. The simulated field discharges are used to compute average drainage rates (DR) for five durations (1-day; 2-day, ..., 5-day). The annual maximum drainage rates for each of the five durations are then fitted to the Gumbel distribution (EV1) for frequency analysis. The return period for an average 3-day (DR3) duration drainage event is estimated, and it is shown that the historical design drainage coefficients (35-50 mm day(-1)) used for surface drainage systems in sugarcane fields along Guyana's coastland have return periods ranging from 1 in 2 years to 1 in 5 years. These return periods are within recommended values commonly used for agricultural drainage systems. (C) 2017 American Society of Civil Engineers.