Divergence of Reference and Measured Evapotranspiration Observations in Sugarcane in Hawai'I, USA.

Standardized reference evapotranspiration (ET) observations and crop specific coefficients are frequently used to assess crop water use in irrigated agriculture. However, reference ET and crop coefficients are not well studied in tropical, humid systems that still require irrigation. Irrigation in these environments may increase with new biofuel and food production, and accurate ET estimates are needed to enhance production sustainability in water limited regions. We measured ET using Eddy Covariance (EC) towers at two well-irrigated sugarcane fields on the leeward (dry) side of Maui, Hawai’i, USA. Both fields were planted with identical sugarcane varieties and used similar drip irrigation and irrigation scheduling and had similar climate with the exception of mean windspeed. We calculated reference ET at the tower sites using several variants of the ASCE/FAO-56 Penman-Monteith equation. We compared the ASCE/FAO equations to a semi-empirical local reference ET derived from weather station data and calibrated to local pan ET (ETpan) and measured ET from the EC towers. Reference ET from the ASCE/FAO approaches exceeded measured ET during the full canopy ET (when crop coefficients suggest measured ET should exceed reference ET). At the windier tower site, the tall ASCE reference ET (ETr) exceeded measured full canopy ET by more than 100% (>800 mm/year) while the ASCE/FAO-56 short surface reference ET (ET0) exceeded full canopy ET by more than 50%. At the less windy site, ETr still exceeded measured full canopy ET by ~40% while ET0 and ETpan were within 10% of each other. The divergence between the various reference ET equations suggests that modifications to the standard reference ET equations may be warranted for irrigated sugarcane in humid regions to better align reference ET calculations with measured and pan ET.