Regional evapotranspiration from a wetland in Central Europe, in a 16-year period without human intervention

Regional evapotranspiration (ET) for Fenéki “Pond” (FP, Kis-Balaton wetland, Hungary) was estimated using weighted canopy areas of five macrophyte classes and open water over an uninterrupted period without any human intervention between 1997 and 2012. In addition to the dominant common reed (Phragmites australis), the cover ratio of the other four plant groups and open water was monitored. The different cover categories of the wetland were separated using images with surface interpretation using the maximum likelihood classification method. Although the annual cover ratio for any class only changed slightly by 1-2%, this wetland was not considered to be in equilibrium, since species turnover (sedge-grass) was observed during this study. Reference ET was calculated by the widely accepted Penman-Monteith (FAO-56) formula from the data of a nearby climate station. Of the five macrophyte classes studied, crop coefficients (as the ratio of measured and reference ET) of three groups were measured locally with a lysimeter. To process the 16-year long ET time series, the means of six-season long Kc values were applied. As the open water of FP was frequently covered by seaweed (Ceratophyllumdemersum), a special evaporation (E) was defined using seaweed settled “A” pans. Compared to evaporation of standard “A” pan, the seasonal average evaporation was 7% higher in “A” pans containing seaweed (C.demersum). Annual and seasonal mean ET totals fluctuated depending on environmental conditions (air temperature) and the type of macrophyte. The range of macrophytes' 16-year annual mean ET varied from 876.8 ± 54.58 mm year−1 (dominant common reed) to 662.8 ± 41.57 mm year−1 (woody deciduous, Salix fragilis, Alnus glutinosa, Populus tremula) with a weighting average of 802.1 ± 50.4 mm year−1 for the regional ET of FP. The average winter ET sum was 17.1 ± 1.19% with a range of 15.7 ± 1.09-20.7 ± 1.43% for common reed and woody deciduous, respectively. The annual mean ET totals were highly correlated with mean daily ET rates of different macrophyte classes. Seasonal daily average ET rates of different macrophyte classes varied between 4.0 ± 0.26 mm day−1 and 2.9 ± 0.19 mm day−1 in common reed and woody deciduous, respectively. The long-term weighted daily mean ET rate for FP was 3.6 ± 0.24 mm day−1. The highest ET/E ratio of 1.09 was for common reed, in an agreement with previous observations, but was lowest in woody deciduous (0.81). In addition to classic statistical approaches, monthly ET data were also used for modeling purposes (ARIMA models). The projected monthly regional ET values supported earlier observed seasonal patterns likely due to peculiarities in wetland water supply. Instead of exclusively using dominant species to estimate regional ET, more accurate results were possible by extending the ET approach to include several macrophyte species. Neglecting the smaller-scale differences in macrophyte, ET could cause more than a 10% overestimation in the regional ET sum. The SARIMA(1,0,0)(0,1,1)12 model contributed positively to the existing information about the ET of the study site. The projected monthly regional ET values retained the earlier observed seasonal pattern until 2020.