Contents lists available at ScienceDirect Agricultural Water Management journal homepage: www.elsevier.com/locate/agwat How significant is the effect of the surface characteristics on the Reference Evapotranspiration estimates? Stavros Alexandris a, *, Nikolaos Proutsos b a Lab of Agricultural Hydraulics, Department of Natural Resources & Agricultural Engineering, Agricultural University of Athens, 75 Iera Odos, 118 55 Athens, Greece b Institute of Mediterranean Forest Ecosystems, Hellenic Agricultural Organization “Demeter”, Terma Alkmanos, 11528 Athens, Greece ARTICLE INFO Keywords: Reference Evapotranspiration ET o Penman-Monteith method Hargreaves-Samani method bare soil evaporation vegetation covered surface ABSTRACT Estimated Reference Evapotranspiration (ET o ), requires the meteorological attributes to be taken above stan- dardized well-watered and vegetation-covered surfaces. However, worldwide such vegetation-reference sites are not common. This results in mistaken estimates of ET o by using the well-known and extensively applied FAO-56 Penman–Monteith formula. Consequently, the use of inappropriate data for ET o estimation from non-ideal surfaces, leads to significant and systematic cumulative errors introducing uncertainties when determining the crop water requirements in a region. Additionally, the existing climatic stations are not spatially distributed but rather concentrated mostly in non-rural urban areas or in local airports, operating above non–standardized surfaces. The purpose of this work is to assess the effect of using daily meteorological data recorded above a well- watered short crop compared to data obtained above dry bare soil, at the same local environment, using two widely used ET o models: the ASCE Penman-Monteith and the Hargreaves-Samani. The results indicate that the meteorological conditions above the different surfaces are quite different, presenting higher temperature and lower relative humidity values above the bare soil surface compared to the well-watered short crop and re- sulting, finally, to errors in the estimation of reference evapotranspiration. These errors appear to enhance as air temperature, vapor pressure deficit, radiation and atmospheric clearness increases, or relative humidity de- creases but are diminishing under adequate soil moisture conditions resulting after rainfall events. The ET o differences vary according to the model each time adopted but the soil substrate influence can be detected by more sophisticated methods, such as FAO56-PM, which consider the energy balance of the surface. 1. Introduction The spatial and temporal variability of essential factors, affecting reference evapotranspiration (ET o ), makes crucial the standardization of the surfaces on which the agrometeorological stations are estab- lished. This variability creates difficulties in the use and accurate per- formance of the proposed ET o estimation equations. ET o is based on potential evapotranspiration from grass or alfalfa crop, which pre- supposes a continuous source of soil moisture and doesn't consider the summer plant dormancy. The ET p is defined by Penman, 1948 as “the amount of water transpired in a given time by a short green crop, completely shading the ground, of uniform height and with adequate water status in the soil profile”. In 1977, FAO proposed a methodology for computing crop evapotranspiration, based in the use of reference evapotranspiration denoted as ET o (Doorenboos and Pruitt, 1977), methodology that re- mains valid at the present day. Twenty-one years later, FAO published a new manual for computing crop water requirements (Allen et al., 1998), that reformulated the concept of "reference evapotranspiration" and espoused the Penman-Monteith equation as a standardized method for ET o estimation. The main factors affecting the ET o include Solar radiation, Air temperature, Relative Humidity and Windspeed. These parameters are related to the nature and properties of the surface above which the measurements are taken and they support the theoretical background of the most reliable model for reference evapotranspiration estimation. The FAO-56 Penman – Monteith equation (Allen et al., 1998) is widely used and considered as a benchmark method for estimating reference evapotranspiration ET o (Walter et al., 2000). It’s data re- quirements, however, are high, since the model demands air tempera- ture, relative humidity, windspeed and solar radiation data. The me- teorological stations measuring these parameters, worldwide, are few and even fewer are the ones that produce “reliable” data (Droogers and https://doi.org/10.1016/j.agwat.2020.106181 Received 12 December 2019; Received in revised form 30 March 2020; Accepted 31 March 2020 ⁎ Corresponding author. E-mail address: stalex@aua.gr (S. Alexandris). Agricultural Water Management 237 (2020) 106181 0378-3774/ © 2020 Elsevier B.V. All rights reserved. T