The development of the methodology for estimating water requirements of a reference crop in urban conditions

VIII, 143 lista

Kratka biografija: list 143. Bibliografija: list. 125-142.

Irrigation

The dissertation presents an analytical framework for defining evapotranspiration as a sustainable mechanism in the urban hydrological cycle. Familiarity with evapotranspiration is important for a large number of hydrological issues in cities. Quantifying evapotranspiration in urban conditions of Niš, as well as the efficiency of existing predictive methods and developing new equations for defining water requirements of reference crops, represent the aims of this study. For the analysis of the set aims, weather data were provided from the following: local automatic stations, evaporators and meteorological observatories in Niš. Six empirical equations were used for the purpose of modeling reference evapotranspiration, the accuracy of which was determined by comparing them with the FAO-56 Penman-Monteith method. The analysis of results has showed that the Priestley and Taylor method exhibits the best performance, but the results have indicated the need for further adaptation of the method to local conditions. The equations were calibrated using optimization methods (genetic algorithm and particle swarm optimization) to best fit the measured data. The sensitivity of the reference evapotranspiration to meteorological parameters was examined in order to achieve the most efficient optimization of the methods. The sensitivity results showed that evapotranspiration is the most sensitive, in defined urban conditions, to changes in solar radiation and maximum air temperature. The performance of the optimized equations was evaluated using statistical indicators that revealed the Priestley and Taylor method and the Stephens and Stewart method as the best optimized methods. By applying multiple linear regression, new equations were developed for the purpose of calculating evapotranspiration with a limited number of data, which has especially highlighted the method based on the ratio of minimum and maximum temperature and solar radiation.

evapotranspiracija, optimizacija, osetljivost, višestruka linearna regresija, jednačine sa ograničenim brojem podataka, urbani uslovi Niša

evapotranspiration, optimization, sensitivity, multiple linear regression, equations with limited data, urban conditions of Niš

551.573:711.522(497.11)(043.3)

Т 220

Tekst