Modeling evapotranspiration from constructed wetlands

The ability to estimate water use by hydrophytes is important in water balance studies of wetlands. Consequently, the objective of this study was to monitor and model ET from constructed wetlands dominated by cattail ( Typha spp.).; In a greenhouse setting, the feasibility of monitoring cattail transpiration with stem flow gauges was investigated. Lack of thermal equilibrium between xylem fluid and sensors on the stem surface coupled with stem heat damage indicated the inappropriateness of installing this type of gauge on cattails.; ET from cattails grown in microlysimeters at College Station and Stephenville, TX, partitioned into about 70% evaporation (E) and 30% transpiration (T) on average during the study period. ET partitioning was estimated by monitoring water E and cattail ET and T gravimetrically. T from individual plants was found to be better represented by subtracting E from ET. ET from cattails growing in constructed wetlands at Stephenville was also quantified using the water budget method.; The crop curves for cattails grown in microlysimeters at College Station during 1998 and 1999 showed comparatively high maximum Kc values (≈1.8). While Kc values for cattails grown in microlysimeters at Stephenville during 1999 showed lower values (<1.5) than those of College Station, cattails grown in larger constructed wetlands at Stephenville showed maximum Kc values of approximately 2. Constructed wetland cattails grew taller than those grown in microlysimeters in both locations due to nonrestricted root growth. Polynomial equations expressing Kc as a function of Julian Day were developed.; Two models for predicting ET from constructed wetlands were developed, calibrated and validated. These models included a crop coefficient model: ET=Kc^*PET where ET is evapotranspiration in mm d−1, Kc is a dimensional cattail crop coefficient, and PET is potential ET in mm d−1, and a parametric model: ET=k1Rs+k2VPD where Rs is solar radiation in MJ m −2 d−1; VPD is vapor pressure deficit in kPa; and k₁, and k ₂, are dimensionless empirical model coefficients.; Model evaluation showed that the crop coefficient model was best in predicting ET.