Introduction:

The FAO-56 Penman-Monteith (PM) is one of the most solid and commonly used methods for estimating reference evapotranspiration (ETo); however, it requires meteorological data that are not always available, so an alternative is the use of reanalysis data.

Objective:

To estimate the error that the NASA-POWER (NP) system data can generate in the ETo of the Comarca Lagunera, Mexico.

Methodology:

Daily and decadal average ETo were estimated in five different ways. In each case, a different method was used to estimate ETo (FAO-56 PM or Hargreaves and Samani [HS]) and a different meteorological data source (measured, NP data or combination of both).

Results:

NP data can be used to provide temperature, solar radiation and relative humidity variables, but not wind speed. The NP data overestimate the measured ETo, an RMSE of 1.15 and 0.89 mm∙d-1 was found for daily and decadal periods, respectively.

Limitations of the study:

A grid error analysis could not be carried out because the number of stations is limited.

Originality:

The use of reanalysis data to estimate ETo has not been analyzed locally.

Conclusion:

When measured data are not available, NP data and the HS equation can be used. When using the FAO-56 PM method and NP data, the in situ wind speed must be available.