Irrigated plantations and their effect on energy fluxes in a semi-arid region of Israel – a validated 3-D model simulation

Abstract

Abstract. A large irrigated biomass plantation was simulated in an arid region of Israel within the WRF-NOAH coupled atmospheric/land surface model in order to assess land surface atmosphere feedbacks. Simulations were carried out for the 2012 summer season (JJA). The irrigated plantations were simulated by prescribing tailored land surface and soil/plant parameters, and by implementing a newly devised, controllable sub-surface irrigation scheme within NOAH. Two model cases studies were considered and compared – Impact and Control. Impact simulates a hypothetical 10 km × 10 km irrigated plantation. Control represents a baseline and uses the existing land surface data, where the predominant land surface type in the area is bare desert soil. Central to the study is model validation against observations collected for the study over the same period. Surface meteorological and soil observations were made at a desert site and from a 400 ha Simmondsia chinensis (Jojoba) plantation. Control was validated with data from the desert, and Impact from the Jojoba. Finally, estimations were made of the energy balance, applying two Penman–Monteith based methods along with observed meteorological data. These estimations were compared with simulated energy fluxes. Control simulates the daytime desert surface 2 m air temperatures (T2) with less than 0.2 °C deviation and the vapour pressure deficit (VPD) to within 0.25 hPa. Desert wind speed (U) is simulated to within 0.5 m s−1 and the net surface radiation (Rn) to 25 W m−2. Soil heat flux (G) is not so accurately simulated by Control (up to 30 W m−2 deviation) and 5 cm soil temperatures (ST5) are simulated to within 1.5 °C. Impact simulates daytime T2 over irrigated vegetation to within 1–1.5 °C, the VPD to 0.5 hPa, Rn to 50 W m−2 and ST5 to within 2 °C. Simulated Impact G deviates up to 40 W m−2, highlighting a need for re-parameterisation or better soil classification, but the overall contribution to the energy balance is small (5–6%). During the night, significant T2 and ST5 cold biases of 2–4 °C are present. Diurnal latent heat values from WRF Impact correspond closely with Penman–Monteith estimation curves, and latent heat magnitudes of 160 W m−2 over the plantation are usual. Simulated plantation sensible heat fluxes are high (450 W m−2) – around 100–110 W m−2 higher than over the surrounding desert. The high relative HFX over the vegetation, driven by high Rn and high surface resistances, indicate that low Bowen ratios should not necessarily be assumed when irrigated plantations are implemented in, and optimized for arid regions. Furthermore, the high plantation T2 magnitudes highlight the importance of considering diurnal dynamics, which drive the evolution of boundary layers, rather than only on daily mean statistics which often indicate an irrigation cooling effect.