Accuracy of reference evapotranspiration (ETo) estimates under data scarcity scenarios in the Iberian Peninsula

14 Pags.- 6 Tabls.- 9 Figs.- Appendix (supplementary data).The standard approach for computing reference crop evapotranspiration (ETo) is the FAO-56 Penman-Monteith (FAO-PM) method, which requires data on air temperature, radiation, air humidity and wind speed. Unlike air temperature the other variables are less frequently available, hindering the application of FAO-PM. A lot of efforts exist to find the best method to estimate FAO-PM ETo when some variables are not available. The FAO-56 manual recommends to estimate the missing variables based on those currently observed (PM-R), or use the less demanding Hargreaves and Samani method (HS). Additionally, if the missing variables are measured at nearby stations, spatial interpolation can be used to estimate the missing data previous to applying FAO-PM (PM-IC). This paper focuses on the comparison, at the monthly time scale, of the performance of these methods to in the Iberian Peninsula. By using 53 weather stations with all data to calculate FAO-PM, data scarcity scenarios are simulated and the mentioned methods are tested (PM-R, HS, PM-IC)PM-IC yielded consistently the best results according to a number of tests. It yielded the lowest mean absolute error (MAE) at 7.56 mm/month, while PM-R yielded values of 10.15 mm/month and HS 9.36 mm/month and biased results. PM-IC was also best at reproducing the long-term variability and trends in ETo. A good and unbiased estimation of monthly ETo time series are required for irrigation planning and crop design.This work was supported by the research projects CGL2011-24185, CGL2011-29263-C02-02, CGL2014-52135-C3-1-R and CGL2014-517221-REDT financed by the Spanish Commission of Science and Technology and FEDER and “LIFE12 ENV/ES/000536-Demonstration and validation of innovative methodology for regional climate change adaptation in the Mediterranean area (LIFE MEDACC)” financed by the EU LIFE program. The first author was supported by the predoctoral FPU program 2013 (Ministerio de Educación, Cultura y Deporte).Peer reviewe