Abstract:
Potential Evapotranspiration (PET) is essential to soil and water studies
and design of hydraulic structures, e.t.c. This study was conducted to evaluate
the relative performance of four commonly used methods of estimating potential
evapotranspiration.
These are: Thornthwaite method (THON), Blaney-Morin-
Nigeria model (BMN), Penman method (PEN) and Penman-Monteith
method
(PM).
Data used for computation of PET were collected from IITA stations. The
PET values were compared on annual, monthly and decadal basis using Pan
evaporation values as reference. The PET values from each of the model was
regressed with Pan values using a linear regression model.
Penman-Monteith closely followed by Blaney-Morin-Nigeria
model were
highly correlated (on annual, monthly and decadal basis) with the observed
values (Epan).
PM overpredicted Epan by about 2.1 % in the humid zone, by
0.42% in the sub-humid environment while BMN overpredicted Epan by about
20%-33%, PEN by about 32%-34%. For THON the range of overprediction was
between 38% and 69% at Onne and Ibadan. The results showed that PM, PEN
andBMN gave comparable estimates of PET. However BMN has the advantage
of fewer numbers of input data required.
While the Thornthwaite method is most affected by temperature, the BMN
is highly sensitive to solar radiation and relative humidity. Simple models relating PET( obtained by each of the method) to Epan
and the relative humidity, were developed.
The models were tested with
independent data sets.
The results of this study are useful in the management
of irrigation
systems, crop simulation models, planning and operation of water resource
projects
