Agricultural University of Athens
Ανάπτυξη υπολογιστικού σχήματος εκτίμησης της ωριαίας εξατμισοδιάπνοης αναφοράς
Abstract
dc:descriptionReference evapotranspiration, being a most important quantity for irrigation, has attracted the intense research effort of numerous scientists. A great number of research papers are being continuously published in this field. Evapotranspiration measurement is costly and requires high quality installation and maintenance, therefore only few measured data are available. Advances in technology have made it easier to acquire measurements (eddy correlation) but the gap between the needed data and the available measurements is still wide. Estimating methods that step in to bridge this gap, providing estimates for various time steps (hourly, daily, monthly, etc.), are still preferred and used by the majority of scientists. These theoretical (ASCE PM) or empirical methods make estimations for various climatic regimes and using various climatic variables. ASCE PM is the most accurate method, suitable for all climatic regimes and requiring four climate variables.The estimation of reference evapotranspiration with limited climatic data is central to the research in this scientific field. It has been approached by a multitude of researchers in a number of ways. In this thesis, a new empirical relation is being proposed for the estimation of hourly reference grass evapotranspiration, using two meteorological variables namely shortwave solar radiation (Rs) and relative humidity (RH). The proposed method is the only method in the literature that estimates hourly values of reference evapotranspiration using only two meteorological variables. The empirical method has been validated with data from semi arid climatic regions (meteorological station of Davis, CA, and the experimental micrometeorological station of the Agricultural University of Athens in Copais, Central Greece). The new empirical method takes into consideration the theoretical daylight duration and the inverse of the natural logarithm of relative humidity and combines them into a single term. Theoretical daylight duration is utilized in weighting incoming solar radiation. The ratio of reference evapotranspiration to incoming solar radiation is larger in summer than in winter. Therefore, bigger ratios correspond to bigger theoretical daylight durations which occur during the warmer part of the year. Formulations accounting for such a weighing have been proposed in the past (Jensen and Haise 1963). The inverse of the natural logarithm of the relative humidity is correlated with the vapor pressure deficit of the atmosphere, which contributes to the aerodynamic term of the PM equation. The correlation was investigated for the calibration and the validation period. A strong correlation was found between the two variables. The theoretical daylight duration and the inverse of the natural logarithm of the relative humidity were combined in a single complex term of the empirical equation. The coefficients of the equation were calculated by the least squares method. The empirical equation has three variables, the incoming solar radiation, the relative humidity, and a complex term that includes the first two variables, the theoretical daylight duration and the inverse of the natural logarithm of the relative humidity. The estimations of the empirical equation were compared with the estimations of the ASCE PM (2005) equation with the use of numerous statistical indices (coefficient of determination, Root Mean Square Error, slope, Index of Agreement).The empirical equation was calibrated with data from the Davis meteorological station of the CIMIS network in CA, USA, from the year 2000, and validated with data from the same station (66911 hourly data for each of the variables). The years for the validation period were selected based on the minimum and maximum values for the precipitation the yearly average wind speed, temperature and the two consecutive years from the calibration year (2000). The empirical equation was also validated with data from the experimental grass reference meteorological station of the Agricultural University of Athens at Copais, Central Greece. For the validation of the equation, 6593 hourly values for each variable were used for the period from Julian day 92 until 366 of 2012. For the whole validation period (66911 hourly values) of Davis station, it was found that the estimations of the empirical equation are in close agreement with the estimations of the ASCE PM (2005) equation. The sum of the hourly values of this period for the ASCE PM equation is equal to 10811 mm, whereas for the empirical equation is equal to 10700 mm or -1.03%. The RMSE for the estimations of the empirical vs the estimations of the ASCE PM was calculated equal to 0.042 mm/h with a minimum value of 0.036 mm/h and a maximum value of 0.045mm/h. The coefficient of determination was found equal to 0.97 for the whole period with a minimum of 0.95 and a maximum value of 1. The slope of the least squares line between the estimations of the empirical formula vs the estimations of the ASCE PM (2005) was found equal to 0.98 for the whole period, with a minimum of 0.95 and a maximum of 1. The Index of Agreement was found equal to 0.991 for the whole period with a minimum of 0.989 and a maximum of 0.998. The estimations of the empirical method and of the ASCE PM for the Davis station were compared for the eight warm periods (DOY 142 until DOY 246, 16913 hourly values), one for each year of the validation period. For the all the warm periods of the validation period, it was found that the empirical method performs satisfactorily when compared with the ASCE PM (2005) method. The cumulative value for the hourly estimations of the ASCE PM (2005) method was found equal to 4616.8 mm while the cumulative value for the hourly estimations of the empirical method was found equal to 4506, which is 2.4% smaller. The coefficient of determination between the estimates of the two methods was found equal to 0.97 for the whole period with a minimum value of 0.97 and a maximum value of 0.98. The RMSE for all the warm periods was found equal to 0.047 mm/h with a minimum value of 0.040 mm/h and a maximum value of 0.055 mm/h. The slope of the least squares line between the estimations of the empirical method and the ASCE PM (2005) method was found equal to 0.98 for the all the eight warm periods with a minimum value of 0.93 and a maximum value of 1. The Index of Agreement was found equal to 0.993 for all the warm periods with a minimum of 0.990 and a maximum of 0.995. The empirical method was also investigated for the experimental grass reference station of the Agricultural University of Athens in the year 2012, DOY 93-366 (6593 hourly values for each variable), and the sum of the hourly estimations of the ASCE PM (2005) was found equal to 1171.3 mm whereas the sum of the hourly estimations of the empirical method was found equal to 1183.8 mm, or 1.1% higher. The coefficient of determination was found equal to 0.97, the RMSE equal to 0.043. the slope of the least squares line between the estimations of the ASCE PM (2005) and the estimations of the empirical method equal to 0.96. The Index of Agreement was found equal to 0.992. Therefore, we can say that the two methods estimate the values of the hourly reference evapotranspiration satisfactorily. The ETo estimation with both methods were grouped according to the magnitudes of vapor pressure deficit and wind speed. RMSE was calculated for each group. Groups from Davis station with VPD>4kPa corresponded to 2.5% of the estimations and had RMSE >0.072 mm/h. For the same station groups with wind speed >7.5 m/s corresponded to 2.1% of the estimations and had RMSE > 0.072mm/h. In the same way for the experimental grass reference station of the Agricultural University of Athens groups with VPD>4.9 corresponded to 0.7% of the estimations and had RMSE>0.072 mm/h. Groups with wind speed >5.7 m/s corresponded to 0.3% of the estimations and had RMSE>0.072 mm/h.The empirical method is applicable to semi arid climates. Its hourly estimations were examined for yearly periods and for the warm periods of each year, and were found satisfactory compared with the estimations of the ASCE PM (2005) method. They were also checked for various ranges of vapor pressure deficit of the atmosphere and wind speed and also found to estimate satisfactorily. We can therefore assume that the aerodynamic term of the ASCE PM equation is adequately represented from the empirical relation.
Degree
thesis:*- Grantor dc:publisher
- Agricultural University of Athens
- Year dc:date
- 2015
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Chatzithomas, Costas
- Χατζηθωμάς, Κωνσταντίνος
Subjects
dc:subject × 11Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/36367