estimation and trend evaluation of reference
TRANSCRIPT
Estimation and trend evaluation of reference evapotranspiration in a humid regionMohammad Valipour1, Mohammad Ali Gholami Sefidkouhi2*, Mojtaba Khoshravesh3
Ital
ian
Jour
nal o
f Agr
omet
eoro
logy
- 1/
2017
Riv
ista
Ita
liana
di A
grom
eteo
rolo
gia
-1/2
017
19
Abstract:In this study, six reference evapotranspiration (ET0) equations were compared and variations of each model have been investigated to detect the main factor(s) of temporal changes of ET0 from 1981 to 2010 in Babolsar, Iran. The results indicate that Priestley-Taylor (PT) estimates ET0 better than other equations during 30-year period. However, performance of all models is different from 2000 to 2008. The study of the variations of air temperature (mean, minimum, and maximum), relative humidity, wind speed, and sunshine hours, underlines that there is a considerable difference between the averages of the meteorological variables during 2000-2008 and the same during the other years. The changes of the wind speed (+1.0 m s-1), sunshine hours (+11.3 hr month-1), average air temperature (+0.7 °C), minimum air temperature (+1.0 °C), difference air temperature (Tmax−Tmin) (−0.6 °C), and relative humidity (−2.7 %) lead to a significant monotonic trend in the confidence level of 99% for these variables as well as ET0 in the study area. During 2000-2008, the average of the wind speed has been increased more than 70 % that it introduces the wind speed as the most important variable for the changes of ET0 and alarms a climatic change in Babolsar.Keywords: climatic change; Iran; Mann-Kendall test; trend analysis; wind speed.
Abstract: In questo studio, sei modelli per la stima dell’evapotraspirazione di riferimento (ET0) sono stati confrontati, al fine di identificare la variabile climatica che ha maggiore effetto sulla variabilità temporale di ET0 nel periodo 1981-2010 in Babolsar, l’Iran. I risultati indicano che il modello di Priestley-Taylor (PT) stima ET0 meglio di altre equazioni, nel trentennio in studio. Tuttavia, le prestazioni di tutti i modelli mostrano qualche anomalia nel periodo 2000-2008. Lo studio delle variazioni termiche (media, minimo e massimo), di umidità relativa, velocità del vento, numero di ore di insolazione, sottolinea che vi è una notevole differenza tra le medie delle variabili climatiche durante 2000-2008 e il resto del trentennio in studio. I cambiamenti di velocità del vento (+1.0 m s-1), numero di ore di insolazione (mese-1 11.3 ore), temperatura media (+0.7 °C), temperatura minima (+1.0 °C), range termico (Tmax-Tmin) (-0.6 °C), e umidità relativa (-2.7%) portano ad una significativa tendenza monotona, con livello di significatività del 99%, per queste variabili e ET0 nell’area di studio. Durante il periodo 2000-2008, la media della velocità del vento è aumentata di oltre il 70%, il che indentifica la velocità del vento come la variabile più importante da cui dipendono le modifiche di ET0 e e in generale i fenomeni di cambiamento climatico in Babolsar.Parole chiave: cambiamento climatico; Iran; test di Mann-Kendall; analisi delle tendenze; velocità del vento.
* Corresponding author’s e-mail: [email protected] Young Researchers and Elite Club, Kermanshah Branch, Islamic Azad University, Kermanshah, Iran2* Department of Water Engineering, Sari Agricultural Sciences and Natural Resources University, Sari, Iran3 Department of Water Engineering, Sari Agricultural Sciences and Natural Resources University, Sari, IranReceived 20April 2016, accepted 25 July 2016
DOI:10.19199/2017.1.2038-5625.019
as temperature-based models with a satisfactory accuracy in humid and other climatic regions:Tabari (2010) compared Makkink (Ma.), Turc (Tu.), PT, and HS in 12 synoptic stations of Iran from 1986 to 2005 and reported better performance of the HS and Tu. rather than the other both models. However, the PT has a higher accuracy in Babolsar. Tabari et al., (2013) compared mass transfer-based models in a humid environment and claimed that the Tr. has highest accuracy in Rasht, Iran. Iramk et al., (2003) investigated radiation-based equations to estimate ET0 in humid regions from 1980 to 1994 and reported the better results of the PT and JH rather than other methods. Valipour (2014a) evaluated 11 mass transfer-based models and concluded that the Pe. and Tr. have the highest accuracy in Iran. Sharifan et al., (2010) compared the Pe., Food and Agriculture Organization (FAO) Penman (FPe.), Hargreaves (Ha.), and FAO Blaney-Criddle (FBC) in Gorgan from 1982 to 2009 and introduced the Pe. as the superior model
1. IntRoductIonThe crop evapotranspiration is the main consumer of water in agriculture. Therefore, an accurate analysis of the variations of evapotranspiration rate may help to better estimate irrigation water requirements and lead significantly to water saving.Among all of radiation, mass transfer, and temperature models for estimating reference evapotranspiration (ET0), there are many studies that recommend Jensen-Haise (JH) and Priestley-Taylor (PT) as radiation-based, Penman (Pe.) and Trabert (Tr.) as mass transfer-based, and Hargreaves-Samani (HS) and Thornthwaite (Th.)
CIANO AGRO 1-17.indb 19 04/04/17 16.42
20
in northwest China during 1994-2010. Zongxing et al., (2014) concluded that decreased wind speed is the main driving force for ET0 decrease in south western China, during 1961-2009 because the higher saturated vapour pressure would restrain evaporation owing to the weakwater vapour exchange under lower wind speed.An accurate survey of the previous studies shows that, in some cases, there are different results in estimation and trend analysis of ET0 due to many reasons.Therefore, the objectives of this study are:(1) Reduction of at least some of the sources of error in previous investigations such as use of insufficient recorded length, use of improper empirical equations in limited weather data conditions, focus on only one of temperature, radiation, or mass transfer-based models, use of improper statistical methods for trend analysis, use of limited statistical indices for evaluation of models error, less attention to variation of error, and less attention to simultaneous study of evapotranspiration and weather parameters variations.For this purpose, suiTab. methods, data, and evaluation indices were employed for estimation of ET0 and evaluation of trend analysis in Babolsar, from 1981 to 2010.(2) Considering the role of different periods to introduce the best model for estimating ET0 on the basis of parameters used in each model.(3) Conducting a sensitive analysis to characterize the values of meteorological variables for which the best performance of each model may be obtained. This not only recommends the use of each model in the best condition, but also helps to apply the results for other regions in the world to justify the international aspect of the study.
2. Methodology And MAteRIAlSFig. 1 and Tab. 1 show the status and values of the annual mean meteorological parameters of Babolsar, respectively.According to World Meteorological Organization (WMO), use of 30-year data series leads to reliable results related to climate (Folland et al., 1999; Ventura et al., 2002). Therefore, in this research, all of the necessary information were collected from Babolsar synoptic station during 1981-2010. In this station, the mean, maximum, and minimum daily air temperature for each month, the mean daily wind speed for each month, the average relative humidity for each month, and the actual duration of sunshine hours for each month were recorded as well as all of them were analyzed by the authors
to estimate ET0 in this region. In other study, Xu and Singh (2001) suggested the use of the Th. method in limited data condition. Sabziparvar and Tabari (2010) compared the Ma., PT, and HS to estimate ET0 in eastern Iran form 1993 to 2005. The authors mentioned that the HS estimates ET0 more accurate than the other models. However, there are also many studies that suggest the use of other temperature/radiation/mass transfer-based models (Azhar et al., 2010; Valipour et al., 2013; Valipour and Montazar 2012; Valipour 2013a,b,c,d; Valipour 2017a,b; Valipour and Gholami Sefidkouhi 2017; Samaras et al., 2014; Valipour 2015d; Vicente-Serrano et al., 2014; Yannopoulos et al., 2015; Zhai et al., 2010).In addition, many investigations analyzed trend of ET0 in the various regions of Iran. Tabari et al., (2012) evaluated spatiotemporal variations of ET0 in arid and semi-arid regions of Iran from 1981 to 2005. Analysis of the impacts of meteorological variables on the temporal trends of ET0 indicated that the increasing trend of ET0 was most likely due to a significant increase in minimum air temperature, while decreasing trend of ET0 was mainly caused by a significant decrease in wind speed. Dinpashoh (2006) claimed that the Caspian Sea shoreline provinces have the lowest ET0 in Iran and south provinces have a rate of ET0 equal to 33 times more than average of ET0 in Iran from 1971 to 2000. However, Hosseinzadeh Talaee et al., (2014) in the newest study reported that the stations located at the southeast, northeast, and northwest corners of Iran record the highest positive change of annual ET0 during 1966-2005.In the other countries, Huntington and Billmire (2014) underlined that the lack of a more consistent increase in ET0, compared with the increases in precipitation and runoff, from 1927 to 2011, was unexpected, but may be explained by various factors including decreasing wind speed, increasing cloudiness, decreasing vapour pressure deficit, and patterns of forest growth in the United States. Jhajharia et al., (2012) reported that the decreases in ET0 are mainly attributed to the net radiation and wind speed, which are also corroborated by the observed trends in these two parameters at almost all the times scales over most of the sites in northeast India during 22 years. Liu and Zhang (2013) showed that the significant decrease in wind speed dominated the change in ET, which offset the effect of increasing air temperature and led to the decrease in ET0 from 1960 to 1993. Increases in air temperature and wind speed together reversed the trend in ET0 and led to the increase in ET0
Ital
ian
Jour
nal o
f Agr
omet
eoro
logy
- 1/
2017
Riv
ista
Ita
liana
di A
grom
eteo
rolo
gia
-1/2
017
CIANO AGRO 1-17.indb 20 04/04/17 16.42
Ital
ian
Jour
nal o
f Agr
omet
eoro
logy
- 1/
2017
Riv
ista
Ita
liana
di A
grom
eteo
rolo
gia
-1/2
017
21
(2)
(3)
(4)
Where, Xi and Yi are the ith observed and estimated values, respectively; X– and Y– are the average of Xi and Yi, and n is the total numbers of data.In addition, trend evaluation was done using regression analysis and Mann-Kendall (MK) test (Mann 1945, Kendall 1975, Gilbert 1987). The related equations for calculating the MK test, statistic S, and the standardized test statistic Z are as follows:
(5)
(6)
(7)
(8)
Where xi and xj are the sequential data values of the time series in the years i and j, n is the length of the time series, tp is the number of ties for the pth value, and g is the number of tied values. Positive values of Z indicate increasing trends, while negative Z values indicate decreasing trends in the time series. The MK tests whether to reject the null hypothesis (H0) and accept the alternative hypothesis (Ha), where H0 is no monotonic trend
(for more information see Islamic Republic of Iran Meteorological Organization (IRIMO) in this link: http://www.irimo.ir/eng/index.php).Although use of lysimeter data reduces uncertainty, however, nowadays FAO Penman Monteith (FPM) model (Allen et al., 1998) is widely applied for estimation of ET0 due to limitation of the lysimeter data (Fan and Thomas 2013; Pejic et al., 2015; Samaras et al., 2014; Shenbin et al., 2006; Valipour 2012a,b,c,d; Valipour 2015a,b,c; Wrachien and Membretti 2015). The FPM model needs to measure many variables, therefore, radiation, temperature, and mass transfer-based models are applied in limited weather data conditions. Tab. 2 shows the models used in this study for estimating ET0 in Babolsar, Iran.In this study, four evaluation indices were selected for comparison of the models as follows:
(1)
Fig. 1 - Status of Babolsar in Iran.Fig. 1 - Stato del Babolsar in Iran.
IcAo code
north latitude
eastlongitude
Altitude (masl)
tmean(°c)
tmax (°c)
tmin (°c)
tmax–tmin (°c)
Relative humidity
(%)
Precipitation (mm/month)
wind Speed (m/s)
Sunshine (hr/month)
40736 36° 43’ 52° 39’ -21 17.5 21.2 13.8 7.4 81.9 77.6 1.6 168.4tab. 1 - Characteristics of Babolsar synoptic station.Tab. 1 - Caratteristiche della stazione sinottica Bablsar.
CIANO AGRO 1-17.indb 21 04/04/17 16.42
According to Fig. 3, the value of the error is more than 10% before 2000 and it is approximately equal to 10% for 2000. Then, the error was reduced to less than 10% after 2000, however, it was again increased from 2009 to 2010. These are confirmed by Fig. 4 in which the lowest error belongs to the period 2000-2008. Fig. 5 shows that the error of the peak points has clearly been reduced during 2000-2008. In addition, Fig.s 6 and 7 indicate that the number of the points in which the error is more than 40%, are too high before 2000, there are no these points during 2000-2008, and the condition is similar to 1981-1999, in the period 2009-2010. Fig. 8 confirms the mentioned cases about an abrupt change after the period of 2000-2008.
22
and Ha is monotonic trend. When |Z|>Z1−α/2, the null hypothesis is rejected and a significant trend exists in the time series. Z1−α/2 is the critical value of Z from the standard normal Tab.
3. Results and discussionsFig. 2 shows performance of the best model (Tr.) for estimating ET0 in each month.As shown, there is a considerable difference between ET0 estimated by Tr. and the same calculated by FPM. Therefore, values of ET0 were estimated for each year during 1981-2010.Figg. 3-7 show some results of estimating ET0 for each year using PT and Tr. models. There is a similar trend for the other models.
Fig. 2 - Comparison between Tr. (estimation) and FPM in Babolsar for each month.Fig. 2 - Confronto fra Tr (stimato) e FPM in Babolsar per ogni mese.
tab. 2 - Models used in this study for estimating ET0.Tab. 2 - Modelli usati nello studio per stimare ET0.
Ital
ian
Jour
nal o
f Agr
omet
eoro
logy
- 1/
2017
Riv
ista
Ita
liana
di A
grom
eteo
rolo
gia
-1/2
017
19-36 valipur.indd 22 11/04/17 14.55
According to Fig. 8, the variations of the evaluation indices are distinguishable from 2000 to 2008. Meanwhile, PT model is the best model for estimating ET0 in Babolsar based on R2 values (the average of R2 values is 0.99 during 1981-2010). Furthermore, Fig. 9 compares the average values of each weather parameter during 2000-2008 versus the same during the other years as well as the minimum and maximum of these two periods.In Fig. 9, although there is a difference between the average values of all weather parameters during 2000-2008 and the same during the other years, all of the average values related to 2000-2008 fall into the variations range (the minimum and maximum values) related to the other years. In the other word, the average values related to 2000-2008, are observable conditions during whole of the 30-years period (they are between the minimum and maximum values of the other years). However, there is a completely different condition for wind
speed; not only the average value of wind speed, during 2000-2008 (2.3 m s-1), is not observed in the other years, but also the maximum value, in the other years, is less than the minimum value from 2000 to 2008 (dashed line). In addition, the average value of wind speed has increased more than 70% during 2000-2008 (from 1.3 m s-1 to 2.3 m s-1). The MK test for all of the variables (Tab. 3) is interpreTab. with respect to the difference between the average values during 2000-2008 and the same during the other years (Fig. 9).According to Tab. 3, there is no monotonic trend for relative humidity and maximum temperature due to the poor changes after 2000 (see also Fig. 9). However, there is a monotonic trend (significant at the confidence level of 99%) for average, minimum, and difference air temperature, wind speed, relative humidity, and sunshine hours owing to considerable differences between two periods (see also Fig. 9).
23
Ital
ian
Jour
nal o
f Agr
omet
eoro
logy
- 1/
2017
Riv
ista
Ita
liana
di A
grom
eteo
rolo
gia
-1/2
017
Fig. 3 - Comparison between PT (estimation) and FPM in Babolsar for each year.Fig. 3 - Confronto fra PT (stimato) e FPM in Babolsar per ogni mese.
CIANO AGRO 1-17.indb 23 04/04/17 16.42
24
In Fig. 10, although there is a difference between the average values of all models during 2000-2008 and the same during the other years, all of the average values related to 2000-2008 fall into
Moreover, Fig. 10 compares the average estimated values of each model during 2000-2008 versus the same during the other years as well as the minimum and maximum of these two periods.
Fig. 4 - Obtained error from the difference between FPM and PT (FPM-PT) for each year.Fig. 4 - Errore ottenuto dalla differenza fra FPM e PT (FPM-PT) per ogni anno.
Variable S Var(s) Z h0 trend difference between two periods
Tmean 212 3140.7 3.8c False Yes DTmean= + 0.7° CTmax 84 3140.7 1.5 True No DTmax= + 0.4° CTmin 297 3141.7 5.3c False Yes DTmin= + 1.0° CTmax-Tmin -222 3139.7 -3.9c False Yes DTmax - Tmin= - 0.6° CRelative Humidity -239 3135.0 -4.3c False Yes DRH= -2,7%Precipitation -41 3141.7 -0.7 True No DP= +4.2 mm/monthWind Speed 182 3140.7 3.2c False Yes DV= + 1.0 m/sSunshine 201 3141.7 3.6c False Yes DS= + 11.3hr/month
a. Significant in confidence level of 90% b. Significant in confidence level of 95% c. Significant in confidence level of 99%
tab. 3 - Results of MK test for weather parameters.Tab. 3 - Risultati del MK test per i parametri meteorologici.
Ital
ian
Jour
nal o
f Agr
omet
eoro
logy
- 1/
2017
Riv
ista
Ita
liana
di A
grom
eteo
rolo
gia
-1/2
017
CIANO AGRO 1-17.indb 24 04/04/17 16.42
the variations range (the minimum and maximum values) related to the other years. In the other word, the average values related to 2000-2008, are observable conditions during whole of the 30-years period (they are between the minimum and maximum values of the other years). However, there is a completely different condition for Tr. model; not only the average value of this model, during 2000-2008 (3.6 mm day-1), is not observed in the other years, but also the maximum value, in the other years, is less than the minimum value from 2000 to 2008 (dashed line). In addition, the average estimated ET0 by this model has increased more than 70% during 2000-2008 (from 2.1 m s-1 to 3.6 m s-1). In addition, it should be noted that ET0 has been increased in all models for the period of 2000-2008. These results confirm the presented interpretations for Fig.s 9, 12, & 13.A regression analysis was also applied in three periods (before and after 2000 as well as during 1981-2010) for better understanding the variations
of each meteorological parameter and to surveying the trends (Fig. 11).Although there are distinguishable trends before and after 2000 for some of the meteorological parameters, variations of wind speed is more considerable. In this case, there is a decreasing trend, after 2000, however, the increasing trend related to before 2000 is too high, and therefore, the total trend (the consequent of the trends during 1981-1999 and 2000-2010) is also increasing. Meanwhile, the value of wind speed decreased drastically during 2009-2010. This confirms the result of Fig.s 3-8 (a similar behaviour between 1981-1999 and 2009-2010 in estimation of ET0 and the evaluation indices) and introduces the wind speed as the most important parameters to control the variations of ET0 in Babolsar. Tab. 4 shows the results of the MK test for evaluating the zero hypotheses for the models used in this study.Although Tab. 4 indicates that there are monotonic trends for all of the models used (except JH), Fig.
25
Ital
ian
Jour
nal o
f Agr
omet
eoro
logy
- 1/
2017
Riv
ista
Ita
liana
di A
grom
eteo
rolo
gia
-1/2
017
Fig. 5 - Comparison between Tr. (estimation) and FPM in Babolsar for each year.Fig. 5 - Confronto fra Tr (stimato) e PFM in Babolsar per ogni anno.
CIANO AGRO 1-17.indb 25 04/04/17 16.42
12 shows that the variations are more considerable for Tr. model.The surveying Tr. model (Fig. 12) indicates that there is a decreasing trend, before 2000, however the increasing trend related to after 2000 is too high, and therefore, the total trend (the consequent of the trends during 1981-1999 and 2000-2010) is also increasing. Note that the length of the first
period (19 years) is more than the length of the second period (11 years), however, the total trend is more affected by the second period (2000-2010). Meanwhile, the maximum value of ET0 estimated by Tr. model, is less that all of the values of this model after 2000. These results are compatible with Fig. 9 (in which the maximum value of the wind speed, in the other years, was less than the
26
Method S Var(s) Z h0 trend
HS 303 3141.7 5.4c False YesJH 23 3141.7 0.4 True NoPe. 213 3141.7 3.8c False YesPT 189 3141.7 3.4c False YesTh. 195 3141.7 3.5c False YesTr. 207 3141.7 3.7c False Yes
FPM 237 3141.7 4.2c False Yesa. Significant in confidence level of 90% b. Significant in confidence level of 95% c. Significant in confidence level of 99%
Fig. 6 - Comparison between relative humidity and FPM/Tr. in Babolsar for each year.Fig. 6 - Confronto fra l’umidità relativa e FPM/Tr in Babolsar per ogni anno.
tab. 4 - Results of MK test for models used.Tab. 4 - Risultati del MK test per i modelli usati.
Ital
ian
Jour
nal o
f Agr
omet
eoro
logy
- 1/
2017
Riv
ista
Ita
liana
di A
grom
eteo
rolo
gia
-1/2
017
CIANO AGRO 1-17.indb 26 04/04/17 16.42
27
Ital
ian
Jour
nal o
f Agr
omet
eoro
logy
- 1/
2017
Riv
ista
Ita
liana
di A
grom
eteo
rolo
gia
-1/2
017
significant in confidence level of 99%, Fig. 13 shows that the ratio of FPM to Tr. model is more considerable.From 1981 to 1999, since the average of wind speed is low (see also Fig. 9), the numerator (calculated values using FPM model) is more than the denominator (estimated values using Tr. model) which led to the values more than 1.0
minimum of the wind speed, during 2000-2008) and underline once again the importance of the wind speed for estimation of ET0 in Babolsar. Tab. 5 shows the results of the MK test for evaluating the zero hypotheses for the ratio of FPM model to models used for estimating ET0 in Babolsar.Although Tab. 5 indicates that the trends of all of the ratios (with the exception of FPM/Pe.) are
Ratio S Var(s) Z h0 trendFPM/HS 209 3141.7 3.7c False YesFPM/JH 263 3141.7 4.7c False YesFPM/Pe. -53 3141.7 -0.9 True NoFPM/PT 185 3141.7 3.3c False YesFPM/Th. 209 3141.7 3.7c False YesFPM/Tr. -189 3141.7 -3.4c False Yes
a. Significant in confidence level of 90% b. Significant in confidence level of 95% c. Significant in confidence level of 99%
Fig. 7 - Comparison between wind speed and FPM/Tr. in Babolsar for each year.Fig. 7 - Confronto fra la velocità del vento e FPM/Tr in Babolsar per ogni anno.
tab. 5 - Results of MK test for ratio of FPM to the models.Tab. 5 - Risultati del MK test per il rapporto fra FPM e i modelli.
CIANO AGRO 1-17.indb 27 04/04/17 16.42
28
Ital
ian
Jour
nal o
f Agr
omet
eoro
logy
- 1/
2017
Riv
ista
Ita
liana
di A
grom
eteo
rolo
gia
-1/2
017
Singh 2016; Valipour et al., 2016; Vicente-Serrano et al., 2014; Zhai et al., 2010), some of different models were applied for estimating ET0 during a specified period without simultaneous evaluation of the weather parameters and finally one of them was introduced as the best model. However, the current study shows that evaluating the weather parameters is necessary for identification of the reason(s) of the variations occurred in the models used for estimation of ET0 and even it leads to changing the best model for the distinguishable period (from 2000 to 2008).Another novelty of this study is determining the best conditions in which each model estimates ET0 with the highest accuracy (Fig. 14) based on the evaluation indices (Equations 1-4). This leads to useable results for other regions in the world.Fig. 14 shows that, for example, the best performance of the PT model (the selected model during 1981-2010) occurs in a wind speed equal to 1.09 m s-1 which belongs to period of the other
and an increasing trend before 2000. From 2000 to 2010, since the average of wind speed is too high (see also Fig. 9), the denominator (estimated values using Tr. model) is more than the numerator (calculated values using FPM model) which led to the values less than 1.0 and a decreasing trend after 2000. The best value for the ratios of Fig. 12 is 1.0 (dashed lines). Therefore, although PT model was introduced as the best model for the 30-year period, the difference of FPM/Tr. with dashed line is high due to variations and trends occurred especially for wind speed after 2000 (Tab.s 3 and 4 as well as Fig.s 2-11) which alarm a climatic change in Babolsar. While the values of FPM/HS and especially FPM/JH are also close to dashed lines. Therefore, JH model is the superior model for estimating ET0 during 2000-2008. This is one of the novelties of the investigation, because in the most previous studies (Ngongondo et al., 2013; Samaras et al., 2014; Temesgen et al., 2005; Thepadia and Martinez 2011; Valipour 2015e,f; Valipour and
Fig. 8 - Variations of evaluation indices in the different models during 1981-2010.Fig. 8 - Variazione degli indici di valutazione per i differenti modelli durante il periodo 1981-2010.
19-36 valipur.indd 28 11/04/17 14.56
29
Ital
ian
Jour
nal o
f Agr
omet
eoro
logy
- 1/
2017
Riv
ista
Ita
liana
di A
grom
eteo
rolo
gia
-1/2
017
Tmin= - 0,6°C), relative humidity (DRH= - 2,7%), and sunshine (DS= + 11.3hr / month), and lack of a significant trend for the maximum temperature (DTmax= +0.4°C) and precipitation (DP= +4.2mm/month). In addition, the regression analysis indicates a considerable difference in the variations of meteorological parameters, before and after 2000, especially for the wind speed in which there is a decreasing trend before 2000 and an increasing trend after 2000. A considerable increasing of the wind speed, equal to more than 70% and a high coefficient for this parameter in Tr. model, led to an increasing trend for variations of Tr. model and a decreasing trend for FPM/Tr. during 2000-2010. According to the results, the wind speed is introduced as the most important factor to control the variations of ET0 in Babolsar. This agrees with Tabari et al., (2011 and 2012) and Kousari and Ahani (2012) in which the wind speed was detected as the
years (Fig. 9), while the best performance of the JH model occurs in a wind speed equal to 1.34 m s-1 which belongs to period of 2000-2008 (Fig. 9). Therefore, the results of this study are confirmed once again using Fig. 14.
4. concluSIonThe PT model (R2=0.99) is the best model for estimating ET0 in Babolsar which agrees with Tabari (2010). Assessment of the variations of ET0 model and evaluation indices underlines a highlighted difference during 2000-2008 and the other years. The results of the MK test and comparison of the annual average of the meteorological parameters before and after 2000, show a significant monotonic trend in confidence level of 99% for the average temperature (DTmean= + 0.7°C), minimum temperature (DTmin= + 1.0° C), wind speed (DV=1.0ms-1), difference temperature (DTmax -
Fig. 9 - Comparison of average data of weather parameters during 2000-2008 versus the other years, the lines show varia-tions range (the minimum and maximum values of each parameter).Fig. 9 - Confronto fra la media dei parametri meteorologici durante il periodo 2000-2008 in confronto con gli altri anni. Le linee mostrano il range di variazione (il valore minimo e massimo di ogni parametro).
CIANO AGRO 1-17.indb 29 04/04/17 16.42
30
Ital
ian
Jour
nal o
f Agr
omet
eoro
logy
- 1/
2017
Riv
ista
Ita
liana
di A
grom
eteo
rolo
gia
-1/2
017
Samaras et al., 2014; Thepadia and Martinez 2011; Valipour 2014b,c; Valipour 2016a,b,c). This investigation can be continued via studying reason(s) of an abrupt change of the wind speed in Babolsar from 2000-2008. Analyzing variations of the meteorological parameters in near stations and provinces and assessment of specific climatic phenomena, in this region, may help to find the reason(s) of this change.As future work, a spatial analysis of the variations of the ET0 may helps us to better understanding the climate condition of the study area.
ReFeRenceSAllen R.G., Pereira L.S., Raes D., Smith M. 1998. Crop
evapotranspiration-Guidelines for computing crop water requirements-FAO Irrigation and drainage paper 56. FAO, Rome, 300, 6541.
main factor for variations of ET0 in Iran. The JH model is selected as the best model for estimating ET0 during 2000-2008, due to increasing wind speed and other meteorological parame ters that led to decreasing accuracy of PT and the other models and this is an alarm for climatic change in Babolsar, Iran.There are three methods to reduce uncertainty of the obtained results: (1) use of more accurate methods to estimate evapotranspiration including lysimeter data (Gowda et al., 2008; Lorite et al., 2012), Bowen ratio (Li et al., 2008; Nagler et al., 2005), eddy covariance (Li et al., 2008; Nagler et al., 2005), remote sensing (Nagler et al., 2005; Rahimi et al., 2015; Sun et al., 2012; Tian et al., 2013), etc. (2) Increasing number of synoptic stations, models, and evaluation indices (3) calibration of ET0 models to localize the coefficients of each model for a certain area (Mallikarjuna et al., 2014;
Fig. 10 - Comparison of average estimated data by evapotranspiration models during 2000-2008 versus the other years, the lines show variations range (the minimum and maximum values).Fig. 10 - Confronto fra la media dei valori di evapotraspirazione utilizzati dai modelli durante il periodo 2000-2008 in con-fronto con gli altri anni. Le linee mostrano il range di variazione (il valore minimo e massimo).
CIANO AGRO 1-17.indb 30 04/04/17 16.42
31
Ital
ian
Jour
nal o
f Agr
omet
eoro
logy
- 1/
2017
Riv
ista
Ita
liana
di A
grom
eteo
rolo
gia
-1/2
017
Fig. 11 - Regression analysis in three periods (before 2000, after 2000, and during 1981-2010) for weather parameters.Fig. 11 - Analisi di regressione nei tre periodi (prima del 2000, dopo il 2000, durante il periodo 1981-2010) per i parametri meteorologici.
CIANO AGRO 1-17.indb 31 04/04/17 16.42
32
Ital
ian
Jour
nal o
f Agr
omet
eoro
logy
- 1/
2017
Riv
ista
Ita
liana
di A
grom
eteo
rolo
gia
-1/2
017
Fig. 12 - Regression analysis in three periods (before 2000, after 2000, and during 1981-2010) for models used.Fig. 12 - Analisi di regressione nei tre periodi (prima del 2000, dopo il 2000, durante il periodo 1981-2010) per i modelli usati.
CIANO AGRO 1-17.indb 32 04/04/17 16.42
33
Ital
ian
Jour
nal o
f Agr
omet
eoro
logy
- 1/
2017
Riv
ista
Ita
liana
di A
grom
eteo
rolo
gia
-1/2
017
ce, SW China, 1961-2004. Climatic Change. 116(2):309-325.
Folland C.K., Miller C., Bader D., Crowe M., Jones P., Plummer N., Richman M., Parker D.E., Rogers J., Scholefield P. 1999. Workshop on Indices and Indicators for Climate Extremes, Asheville, NC, USA, 3-6 June 1997 Breakout Group C: Temperature Indices for Climate Extremes. Climatic Chang. 42(1):31-43.
Gilbert R.O. 1987. Statistical methods for environ-
Azhar A.H., Perera B.J.C. 2010. Evaluation of re-ference evapotranspiration estimation methods under Southeast Australian conditions. J Irrig Drain Eng. 137(5):268-279.
Dinpashoh Y. 2006. Study of reference crop evapo-transpiration in IR of Iran. Agr Water Manage. 84(1):123-129.
Fan Z.X., Thomas A. 2013. Spatiotemporal varia-bility of reference evapotranspiration and its contributing climatic factors in Yunnan Provin-
Fig. 13 - Regression analysis in three periods (before 2000, after 2000, and during 1981-2010) for ratio of FPM to the models.Fig. 13 - Analisi di regressione nei tre periodi (prima del 2000, dopo il 2000, durante il periodo 1981-2010) per il rapporto fra FPM e i modelli.
CIANO AGRO 1-17.indb 33 04/04/17 16.42
34
Ital
ian
Jour
nal o
f Agr
omet
eoro
logy
- 1/
2017
Riv
ista
Ita
liana
di A
grom
eteo
rolo
gia
-1/2
017
Jensen M.E., Haise H.R. 1963. Estimation of eva-potranspiration from solar radiation. J Irrig Drain Div. 89:15-41.
Jhajharia D., Dinpashoh Y., Kahya E., Singh V.P., Fakheri-Fard A. 2012. Trends in reference evapotranspiration in the humid region of northeast India. Hydrol Process. 26(3):421-435.
Kendall M.G. 1975. Rank Correlation Methods, 4th edition, Charles Griffin, London.
Kousari M.R., Ahani H. 2012. An investigation on reference crop evapotranspiration trend from 1975 to 2005 in Iran. Int J Climatol. 32(1): 29-43.
Li S., Kang S., Zhang L., Li F., Zhu Z., Zhang B. 2008. A comparison of three methods for determining vineyard evapotranspiration in the arid desert regions of northwest China. Hydrol Process. 22(23):4554-4564.
Liu X., Zhang D. 2013. Trend analysis of reference evapotranspiration in Northwest China: The roles of changing wind speed and surface air temperature. Hydrol Process. 27(26):3941-3948.
mental pollution monitoring. John Wiley & Sons. Bahman 26, 1365 AP - Nature - 336 pages. ISBN 0471288780, 9780471288787
Gowda P.H., Chavez J.L., Colaizzi P.D., Evett S.R., Howell T.A., Tolk J.A. 2008. ET0 mapping for agricultural water management: present status and challenges. Irrig Sci. 26(3):223-237.
Hargreaves G.L., Samani Z.A. 1985. Reference crop evapotranspiration from temperature. Appl Eng Agr. 1(2):96-99.
Hosseinzadeh Talaee P., Some’e B.S., Ardakani S.S. 2014. Time trend and change point of reference evapotranspiration over Iran. Theore Appl Climatol. 116(3-4):639-647.
Huntington T.G., Billmire M. 2014. Trends in Precipitation, Runoff, and Evapotranspiration for Rivers Draining to the Gulf of Maine in the United States. J Hydrometeorol. 15(2):726-743.
Irmak S., Irmak A., Allen R.G., Jones J.W. 2003. Solar and net radiation-based equations to estimate reference evapotranspiration in humid climates. J Irrig Drain Eng. 129(5):336-347.
Fig. 14 - The values of the meteorological variables for the best performance of the models based on the evaluation indices, the parentheses show the standard deviation.Fig. 14 - Valori delle variabili meteorologiche per la migliore performance dei modelli, basata sugli indici di valutazione. Fra parentesi i valori della deviazione standard.
CIANO AGRO 1-17.indb 34 04/04/17 16.42
35
Ital
ian
Jour
nal o
f Agr
omet
eoro
logy
- 1/
2017
Riv
ista
Ita
liana
di A
grom
eteo
rolo
gia
-1/2
017
Sun Z., Gebremichael M., Ardo J., Nickless A., Caquet B., Merboldh L., Kutschi W. 2012. Estimation of daily evapotranspiration over Africa using MODIS/Terra and SEVIRI/MSG data. Atmos Res. 112:35-44.
Tabari H. 2010. Evaluation of Reference Crop Evapotranspiration Equations in Various Climates. Water Resour Manage. 24(10):2311-2337.
Tabari H., Aeini A., Talaee P.H., Some’e BS. 2012. Spatial distribution and temporal variation of reference evapotranspiration in arid and semi-arid regions of Iran. Hydrol Process. 26(4):500-512.
Tabari H., Grismer M.E., Trajkovic S. 2013. Comparative analysis of 31 reference evapo-transpiration methods under humid conditions. Irrig Sci. 31(2):67-80.
Tabari H., Marofi S., Aeini A., Talaee P.H., Moham-madi K. 2011. Trend analysis of reference evapo- transpiration in the western half of Iran. Agr Forest Meteorol. 151(2):128-136.
Thornthwaite C.W. 1948. An approach toward a rational classification of climate. Geogr Rev. 38:55-94.
Temesgen B., Eching S., Davidoff B., Frame K. 2005. Comparison of some reference evapotranspiration equations for California. J Irrig Drain Eng. 131(1):73-84.
Tian F., Qiu G., Yang Y., Lu Y., Xiong Y. 2013. Estimation of evapotranspiration and its partition based on an extended three-temperature model and MODIS products. J Hydrol. 498:210-220.
Trabert W. 1896. Neue Beobachtungen uber Ver-dampfungsgeschwindigkeiten. Meteorol Z 13:261-263.
Valipour M. 2012a. Number of Required Observation Data for Rainfall Forecasting According to the Climate Conditions. Am J Sci Res. 74:79-86.
Valipour M. 2012b. Hydro-module determination for Vanaei Village in Eslam Abad Gharb, Iran. Arpn j Agric Biol Sci. 7(12):968-976.
Valipour M. 2012c. Ability of Box-Jenkins Mo-dels to Estimate of Reference Potential Evapotranspiration (A case Study: Mehrabad Synoptic Station, Tehran, Iran). IOSR J Agri Veter Sci. (IOSR-JAVS) 1(5):1-11.
Valipour M. 2012d. A Comparison between Ho-rizontal and Vertical Drainage Systems (Include Pipe Drainage, Open Ditch Drainage, and Pumped Wells) in Anisotropic Soils. IOSR J Mech Civil Eng. (IOSR-JMCE). 4(1):7-12.
Valipour M. 2014a. Application of new mass transfer formulae for computation of evapotranspiration. J Appl Water Eng Res. 2(1):33-46.
Valipour M. 2014b. Use of average data of 181 synoptic stations for estimation of reference crop
Lorite I.J., Santos C., Testi L., Fereres E. 2012. Design and construction of a large weighing lysimeter in an almond orchard. Spanish J Agr Res. 10(1):238-250.
Mallikarjuna P., Jyothy S.A., Murthy D.S., Reddy K.C. 2014. Performance of Recalibrated Equations for the Estimation of Daily Reference Evapotranspiration. Water Resour Manage. 28(13):4513-4535.
Mann H.B. 1945. Non-parametric tests against trend, Econometrica 13:163-171.
Pejic B., Aksic M., Mackic K., Sekularac G. 2015. Response of Potato to Water Stress in Southern Serbia. Austin J. Irrig. 1(1): 1001. In Press. Asses-sed: 23 October 2015.
Penman H.C. 1948. Natural evaporation from open water, bare soil and grass. Proc R Soc Lond Ser A 193:120-145.
Nagler P.L., Scott R.L., Westenburg C., Cleverly J.R., Glenn E.P., Huete A.R. 2005. Evapotranspiration on western US rivers estimated using the Enhanced Vegetation Index from MODIS and data from eddy covariance and Bowen ratio flux towers. Remot Sens Environ. 97(3):337-351.
Ngongondo C., Xu C.Y., Tallaksen L.M., Alemaw B. 2013. Evaluation of the FAO Penman–Montheith, Priestley–Taylor and Hargreaves models for estimating reference evapotranspiration in southern Malawi. Hydrol Res. 44(4):706-722.
Priestley C.H.B., Taylor R.J. 1972. On the assessment of surface heat flux and evaporation using large-scale parameters. Month Weather Review. 100(2):81-92.
Rahimi S., Gholami Sefidkouhi M.A., Raeini-Sarjaz M., Valipour M. 2015. Estimation of actual evapotranspiration by using MODIS images (a case study: Tajan catchment). Arch Agron Soil Sci. 61(5):697-709.
Sabziparvar A.A., Tabari H. 2010. Regional estimation of reference evapotranspiration in arid and semiarid regions. J Irrig Drain Eng. 114(2):198-203.
Samaras D.A., Reif A., Theodoropoulos K. 2014. Evaluation of radiation-based reference evapo- transpiration models under different Medi-terranean climates in central Greece. Water Resour Manage. 28(1):207-225.
Sharifan H., Dehghani A.A., Karimirad I. 2010. Correction Factor for Hargreaves-Samani method to Estimate ETo (Case Study: Gorgan Synoptic Station). J Water Soil Conserv. 19(3):227-236. (in Persian with English Abstract).
Shenbin C., Yunfeng L., Thomas A. 2006. Climatic change on the Tibetan Plateau: potential evapotranspiration trends from 1961-2000. Climatic Change. 76(3-4):291-319.
CIANO AGRO 1-17.indb 35 04/04/17 16.42
36
Ital
ian
Jour
nal o
f Agr
omet
eoro
logy
- 1/
2017
Riv
ista
Ita
liana
di A
grom
eteo
rolo
gia
-1/2
017
and its Solutions, Kish Island, Iran. Assessed 20 October 2013.
Valipour M., Montazar A.A. 2012. An Evaluation of SWDC and WinSRFR Models to Optimize of Infiltration Parameters in Furrow Irrigation. Am J Sci Res. 69:128-142.
Valipour M., 2017a. Status of land use change and irrigation in Europe by 2035 and 2060. J Water Land Dev. Accepted. In Press.
Valipour M., 2017b. Global experience on irrigation management under different scenarios. J. Water Land Dev. Accepted. In press.
Valipour M., Gholami Sefidkouhi M.A. 2017. Temporal analysis of reference evapotranspiration to detect variation factors. Int J Glob Warm. doi: 10.1504/IJGW.2018.10002058
Valipour M., Gholami Sefidkouhi M.A., Raeini-Sarjaz M. 2016. Selecting the best model to estimate potential evapotranspiration with respect to climate change and values of cold and warm seasons. Agric Water Manage. 180(Part A):50-60.
Ventura F., Rossi Pisa P., Ardizzoni E. 2002. Tem-perature and precipitation trends in Bologna (Italy) from 1952 to 1999. Atmos Res. 61(3):203-214.
Vicente-Serrano S.M., Azorin-Molina C., Sanchez-Lorenzo A, Revuelto J, Lopez-Moreno JI, Gonzalez-Hidalgo JC, Moran-Tejeda E, Espejo F. 2014. Reference evapotranspiration variability and trends in Spain, 1961-2011. Global Planet Change. 121:26-40.
Wrachien D.D., Mambretti S. 2015. Irrigation and Drainage Systems in Flood-prone Areas: The Role of Mathematical Models. Austin J. Irrig. 1(1): 1002. In Press. Assessed: 25 November 2015.
Xu C.Y., Singh V.P. 2001. Evaluation and gene-ralization of temperature-based methods for calculating evaporation. Hydrol Process. 15(2):305-319.
Yannopoulos, S.I., Lyberatos, G., Theodossiou, N., Li, W., Valipour, M., Tamburrino, A., Angelakis, A.N., 2015. Evolution of Water Lifting Devices (Pumps) over the Centuries Worldwide. Water 7 (9), 5031-5060.
Zhai L., Feng Q., Li Q., Xu C. 2010. Comparison and modification of equations for calculating evapotranspiration (ET) with data from Gansu province, northwest China. Irrig Drain. 59(4):477-490.
Zongxing L., Qi F., Wei L., Tingting W., Yan G., Yamin W., Aifang C., Jianguo L., Li L. 2014. Spatial and temporal trend of potential evapotranspiration and related driving forces in Southwestern China, during 1961-2009. Quater Int. doi:10.1016/j.quaint.2013.12.045
evapotranspiration by temperature-based methods. Water Resour Manage. 28(12):4237-4255.
Valipour M. 2014c. Analysis of potential evapo-transpiration using limited weather data. Appl Water Sci. doi: 10.1007/s13201-014-0234-2
Valipour M. 2013a. Use of surface water supply index to assessing of water resouces management in Colorado and Oregon, US, Adv Agric Sci Eng Res. 3(2):631-640
Valipour M. 2013b. Necessity of Irrigated and Rain-feld Agriculture in the World. Irrig Drain Sys Eng. e001.
Valipour M. 2013c. Increasing irrigation efficiency by management strategies: cutback and surge irri-gation. Arpn J. Agric Biol Sci. 8(1):35-43
Valipour M. 2013d. Evolution of Irrigation-Equipped Areas as Share of Cultivated Areas. Irrig Drain Sys Eng. 2(1):e 114.
Valipour M. 2015a. Importance of solar radiation, temperature, relative humidity, and wind speed for calculation of reference evapotranspiration. Arch Agron Soil Sci. 61(2):239-255.
Valipour M. 2015b. Future of agricultural water management in Africa. Arch Agron Soil Sci. 61(7):907-927.
Valipour M. 2015c. Study of different climatic conditions to assess the role of solar radiation in reference crop evapotranspiration equations. Arch Agron Soil Sci. 61(5):679-694.
Valipour M. 2015d. Land use policy and agricultural water management of the previous half of century in Africa. Appl Water Sci. 5(4):367-395.
Valipour M. 2015e. Comparative Evaluation of Radiation-Based Methods for Estimation of Potential Evapotranspiration. J Hydrol Eng. 20(5):04014068.
Valipour M. 2015f. Investigation of Valiantzas’ evapotranspiration equation in Iran. Theor Appl Climatol. 121(1−2):267-278.
Valipour M. 2016a. How Much Meteorological Information Is Necessary to Achieve Reliable Accuracy for Rainfall Estimations? Agriculture. 6(4):53.
Valipour M. 2016b. How do different factors impact agricultural water management? Open Agric. 1(1):89-111.
Valipour M. 2016c. Variations of land use and irrigation in Americas for next decades under different scenarios. Irriga (Brazilian Journal of Irrigation and Drainage). 1(1):262-288.
Valipour M., Mousavi S.M., Valipour R., Rezaei E. 2013. A New Approach for Environmental Crises and its Solutions by Computer Modeling. The 1st International Conference on Environmental Crises
CIANO AGRO 1-17.indb 36 04/04/17 16.42