Geophysical Research Abstracts

Vol. 17, EGU2015-8556-1, 2015

EGU General Assembly 2015
Author(s) 2015. CC Attribution 3.0 License.

Sensitivity analysis of various Potential Evapotranspiration formulas for

Crete Island in Greece
Spyridon Paparrizos (1), Fotios Maris (2), Orestis Papageorgiou (3), Konstantinos Karagiorgos (4), Sven Fuchs
(4), and Andreas Matzarakis (1)
(1) Faculty of Environment and Natural Resources, Albert-Ludwigs University of Freiburg, D-79085 Freiburg, Germany
(spipap@gmail.com), (2) Department of Forestry and Management of the Environment and Natural Resources Democritus
University of Thrace, Pantazidou str. 192, GR-68200 Orestiada, Greece (fmaris@fmenr.duth.gr), (3) Department of
Mathematics, University of Athens, Panepistimiopolis, GR-15784 Athens, Greece (o.papageorgiou@windowslive.com), (4)
University of Natural Resources and Life Sciences, Institute of Mountain Risk Engineering, Vienna, Austria
(kkaragio@boku.ac.at)

Evapotranspiration is an important component of the hydrologic cycle as it can significantly affect the water
budget of the natural (i.e. approximately 62% of all precipitation falling on land is evapotranspirated). In the
current study, a sensitivity analysis of various PET formulas was performed for the Crete Island in southern
Greece in order to evaluate the influence of the various coefficients in the estimation of PET. Results indicated that
some PET formulas perform more accurate results than others. Specifically, the PET equation after Jensen-Haise
(PETJen) that is based on mean air temperature (Tmean) and solar radiation (Rs) input data and the 1st version of
Hamon (PETHam1) that uses min (Tmin), mean (Tmean), max (Tmax) air temperature data and Day Length (DL)
input data during the sensitivity analysis indicated that they can be used in order to estimate with high accuracy
the potential evapotranspiration in Crete Island.

Sensitivity analysis of various Potential Evapotranspiration

formulas for Crete Island in Greece
Spyridon Paparrizos1, Fotios Maris2, Orestis Papageorgiou3, Konstantinos Karagiorgos4, Sven Fuchs4, Andreas Matzarakis1
Abstract
Evapotranspiration is an important component of the hydrologic cycle as it can
significantly affect the water budget of the natural (i.e. approximately 62% of all
precipitation falling on land is evapotranspirated). In the current study, a sensitivity
analysis of various PET formulas was performed for the Crete Island in southern
Greece in order to evaluate the influence of the various coefficients in the
estimation of PET. Results indicated that some PET formulas perform more
accurate results than others. Specifically, the PET equation after Jensen-Haise
(PETJen) that is based on mean air temperature (Tmean) and solar radiation (Rs)
input data and the 1st version of Hamon (PETHam1) that uses min (Tmin), mean
(Tmean), max (Tmax) air temperature data and Day Length (DL) input data during the
sensitivity analysis indicated that they can be used in order to estimate with high
accuracy the potential evapotranspiration in Crete Island.
Introduction
Potential evapotranspiration (PET) is defined as the amount of water that can
potentially evaporate and transpire from a vegetated surface with no restrictions
other than the atmospheric demand. Knowledge of PET rates is essential for a
variety of applications, including hydrological modeling, irrigation planning,
agricultural water use, geo-botanical studies and estimation of sensitive-to-climatic
change, aridity indices (Paparrizos et al. 2014). Sensitivity analysis investigates
the effect of change of one factor on another. In many cases it is usually the first
step towards model calibration because it answers several questions such as what
degree of care should be taken for parameter estimation and which is the relative
importance of various parameters (Cho and Lee, 2001).
Research area and Methods
Crete Island is located in the southern part of Greece. It has an area of 8.342 km2,
total population of 623.065 inhabitants and the main occupation of population is
employed in the primary section (Matzarakis and Nastos, 2011; Bleta et al., 2014).
The mean annual precipitation is 822 mm and mean annual Tmean 17.2 oC. EmPEst
software (Kostinakis et al., 2011) and RayMan (Matzarakis et al., 2007; 2010) were
applied in order to estimate the PET formulas. ArcGIS 10.2.1 was used for
mapping the characteristics of the study area. Additionally, sensitivity analysis and
the beanplots (Muthers and Matzarakis, 2010) were performed using Mathworks
2014a. Figure 1 depicts the characteristics of the study area along with the
meteorological stations that were used in the current study for the years 19812000. Table 1 presents the PET formulas that were used in order to perform the
sensitivity analysis.

Common
name
Hargreaves
Hamon 1
Hamon 2
Hamon 3
Romanenko
Caprio
JensenHaise
Turc
ASCE

Note: RH = Relative air humidity (%); Tmean = mean air temperature (oC); Tmax = maximum air temperature (oC); Tmin = minimum
air temperature (oC); Tdew = dew point temperature (oC); Rs = total global solar radiation (MJ m-2); Ra = total solar radiation
reaching the atmosphere's surface (MJ m-2); Rn = total net solar radiation (MJ m-2); WS = wind speed (m s-1); = latent heat of
vaporization (MJ kg-1); = proportionally coefficient (unitless - set as 1.2); SVD = saturated vapour density at mean air
temperature (g m-3); = the slope of saturation vapor pressure curve at air temperature T (kPa oC -1); G = the soil heat flux
density at the crop surface MJ m-2; for daily periods the value is 0); U2 = the mean wind speed at 2 m height (m s-1); = the
psychrometric constant (kPa oC -1); es = the mean saturation vapor pressure at 1.5-2.5 m height (kPa); ea = the mean actual vapor
pressure at 1.5-2.5 m height (kPa); and Cn = a numerator constant that changes with reference type. Cn takes the value 900 (K
mm s3 Mg-1) for short and tall crops, respectively. Cd = a denominator constant that changes with reference type and takes the
values of 0.34 (s m-1) and 0.38 (s m-1) for short and tall crops, respectively.

Tmean
PETHar

18.2
0.2
99.3
5.7
3.6
51.8
53.5
40.9
< 0.01

PETHam1
PETHam2
PETHam3
PETRom
PETCap
PETJen
PETTur
PETASCE

Results and Discussion

9 PETref formulas were used for the estimation of Potential Evapotranspiration in
Crete Island. Through the sensitivity analysis, the formulas that described more
accurate the results were PETHam1 and PETJen equations. According to Table 2, the
results of the PETHam1 and PETJen are very similar to the average results. These 2
equations use Tmean, Tmin Tmax, DL and Rs as input data. Especially for the PETJen
equation, the sensitivity analysis indicated that Tmean and Rs which are the input
data of the equation have almost the same effect on the equation (Tmean 53,5%
while Rs 46.5%). These results constitute PETHam1 and PETJen as the most suitable
equations for the estimation of Potential Evapotranspiration over Crete Island in
Greece. Therefore, the estimation of potential evapotranspiration is influenced
most strongly by air temperature and solar radiation. Additionally, PETRom, PETCap
and PETASCE appeared in the beanplots analysis in figure 2 with greater range,
compared with PETJen and PETHam1. Finally, according to the beanplots, it is
obvious the values for the PETJen and PETHam1 follow a better distribution
compared with the other PET formulas.

Mathematical equation

PETASCE 2.9

Tmin

Tmax

Tdew

Crete Island
Total results
Wind speed

32.3 48.7
19.4 29.8

DL

Rs

Ra

50.6
0.7
0.8

38.2 55.4
57.1 31.7

7.5
48.2
46.5
59.1

2.9

2.3

< 0.01

< 0.01

94.8

PETJen

59.1
53.5

PETCap

46.5

51.8
3.6

48.2
57.1

5.7

31.7

38.2

PETHam1 0.2
PETHar

19.4
18.2

10

20

29.8

50.6

32.3

48.7

30
Tmin

0.7

40
Tmax

Tdew

50
RH

Wind speed

60

70
DL

Rs

Ra

Bleta A., Nastos P. and Matzarakis A., 2014. Assessment of bioclimatic conditions on Crete Island, Greece. Regional Environmental Change 14, 1967-1981.
Cho S.M. and Lee M.W., 2001. Sensitivity considerations when modelling hydrologic processes with digital elevation model. Journal of the American Water Resources Association 37(4), 931934.
Kostinakis K., Xystrakis F., Theodoropoulos K., Stathis D., Eleftheriadou E. and Matzarakis A., 2011. Estimation of reference potential evapotranspiration with focus on vegetation science the EmPEst software. Journal of Irrigation and Drainage Engineering 137, 616619.
Matzarakis A., Rutz F. and Mayer H., 2007. Modeling radiation fluxes in simple and complex environmentsApplication of the RayMan model. International Journal of Biometeorology 51(4), 323334.
Matzarakis A., Rutz F. and Mayer H., 2010. Modelling radiation fluxes in simple and complex environments: basics of the RayMan model. International Journal of Biometeorology 54(2), 131-139.
Matzarakis A. and Nastos P., 2011. Analysis of tourism potential for Crete Island, Greece, Global NEST Journal 13(2), 141-149.
Muthers S. and Matzarakis A., 2010. Use of beanplots in applied climatology - A comparison with boxplots, Meteorologische Zeitschrift, 6, 639-642.
Paparrizos S., Maris F. and Matzarakis A., 2014. Estimation and Comparison of Potential Evapotranspiration based on daily and monthly data from Sperchios River valley in Central Greece, Global NEST Journal 16(2), 204-217.

April 2015

Department of Forestry and Management of the Environment and

Natural Resources Democritus University of Thrace, Pantazidou str.
192, GR-68200 Orestiada, Greece, email: fmaris@fmenr.duth.gr

0.8

99.3

References

Faculty of Environment and Natural Resources, Albert-Ludwigs

University of Freiburg, D-79085 Freiburg, Germany, email:
spipap@gmail.com , andreas.matzarakis@meteo.uni-freiburg.de

7.5

55.4

PETHam2

Tmean

1248.9
1303.9
879.4
1891.5
1033.7
1088.4
1030.9
1033.9
1190.6

94.8
40.9

PETHam3

PET(mm)
1048.3

0.7

2.3

PETTur

PETRom

Rn

Department of Mathematics, University of Athens, Panepistimiopolis, GR-15784

Athens, Greece, email: o.papageorgiou@windowslive.com

University of Natural Resources and Life Sciences, Institute of Mountain Risk

Engineering, Vienna, Austria, email: kkaragio@boku.ac.at , sven.fuchs@boku.ac.at

0.7
80

Rn

90

100