Optimal Interpolation scheme to generate reference crop evapotranspiration

Miquel Tomas-Burguera, Santiago Beguería, Sergio Vicente-Serrano, Marco Maneta

Research output: Contribution to journalArticlepeer-review

15 Scopus citations


We used an Optimal Interpolation (OI) scheme to generate a reference crop evapotranspiration (ETo) grid, forcing meteorological variables, and their respective error variance in the Iberian Peninsula for the period 1989–2011. To perform the OI we used observational data from the Spanish Meteorological Agency (AEMET) and outputs from a physically-based climate model. To compute ETo we used five OI schemes to generate grids for the five observed climate variables necessary to compute ETo using the FAO-recommended form of the Penman-Monteith equation (FAO-PM). The granularity of the resulting grids are less sensitive to variations in the density and distribution of the observational network than those generated by other interpolation methods. This is because our implementation of the OI method uses a physically-based climate model as prior background information about the spatial distribution of the climatic variables, which is critical for under-observed regions. This provides temporal consistency in the spatial variability of the climatic fields. We also show that increases in the density and improvements in the distribution of the observational network reduces substantially the uncertainty of the climatic and ETo estimates. Finally, a sensitivity analysis of observational uncertainties and network densification suggests the existence of a trade-off between quantity and quality of observations.

Original languageEnglish
Pages (from-to)202-219
Number of pages18
JournalJournal of Hydrology
StatePublished - May 2018


  • Optimal Interpolation
  • Reference crop evapotranspiration database
  • Regional Climate Model
  • Uncertainty estimation
  • Weather stations


Dive into the research topics of 'Optimal Interpolation scheme to generate reference crop evapotranspiration'. Together they form a unique fingerprint.

Cite this