Actual and Reference Evaporative Losses and Surface Coefficients of a Maize Field during Nongrowing (Dormant) Periods

Effective water resources planning, allocation, management, and use in agroecosystems require accurate quantification of actual evapotranspiration $\left({\mathrm{ET}}_c\right)$ during growing and nongrowing (dormant) periods. Prediction of ${\mathrm{ET}}_c$ for a variety of vegetation surfaces during the growing season has been researched extensively, but relatively little information exists on evaporative losses during nongrowing periods for different surfaces. The objectives of this research were to evaluate ${\mathrm{ET}}_c$ in relation to available energy, precipitation, and grass and alfalfa-reference ET ( ${\mathrm{ET}}_o$ and ${\mathrm{ET}}_r$ ) for a maize (Zea mays. L) field and to analyze the dynamics of surface coefficients $\left(K_c\right)$ during the nongrowing period (October 15–April 30). The evaporative losses were measured using a Bowen ratio energy balance system (BREBS) on an hourly basis and averaged over $24\mathrm{h}$ for three consecutive nongrowing periods: 2004–2005 (Season I), 2005–2006 (Season II), and 2006–2007 (Season III). BREBS-measured ${\mathrm{ET}}_c$ was approximately 50% of available energy ( $R_n-G$ ; $R_n$ is net radiation and $G$ is soil heat flux density) during normal and wet seasons (Seasons I and III) and 41% of available energy during a dry season (Season II). Cumulative ${\mathrm{ET}}_c$ ranged from $133\mathrm{mm}$ in Season II to $167\mathrm{mm}$ in Season III and exceeded precipitation by 21% during the dry season. The ratio of ${\mathrm{ET}}_c$ to precipitation was 0.85 in Season I, 1.21 in Season II, and 0.41 in Season III. ${\mathrm{ET}}_c$ was approximately 50% of ${\mathrm{ET}}_o$ and 36% of ${\mathrm{ET}}_r$ in both Seasons I and III, whereas in Season II, ${\mathrm{ET}}_c$ was 32% of ${\mathrm{ET}}_o$ and 23% of ${\mathrm{ET}}_r$ . Overall, measured ${\mathrm{ET}}_c$ during the dormant season was generally most strongly correlated with radiation terms, particularly $R_n$ , albedo, incoming shortwave radiation, and outgoing longwave radiation. Average surface coefficients over the three seasons were 0.44 and 0.33 for grass and alfalfa-reference surfaces, respectively. Using geometric mean $K_c$ values to calculate ${\mathrm{ET}}_c$ using a $K_c{\mathrm{ET}}_{\mathrm{ref}}$ approach over the entire nongrowing season yielded adequate predictions with overall root mean square deviations of 0.64 and $0.67\mathrm{mm}{\text{day}}^{-1}$ for ${\mathrm{ET}}_o$ and ${\mathrm{ET}}_r$ , respectively. Estimates of ${\mathrm{ET}}_c$ using a dual crop coefficient approach were good on a seasonal basis, but performed less well on a daily basis. Regression equations that were developed (accounting for serial autocorrelation in the ${\mathrm{ET}}_c$ and ${\mathrm{ET}}_{\mathrm{ref}}$ time series) yielded good estimates of ${\mathrm{ET}}_c$ . Considering nongrowing period evaporative losses in water budget calculations would enable water regulatory agencies to better account for water use in hydrologic balance calculations over the entire year rather than only for the growing season and to better assess the progression and availability of water resources for the next growing season.