Karadeniz Bölgesi İklim Koşullarına Uygun Jensen Haise Yöntemi Referans Evapotranspirasyon Tahmin Eşitliklerinin Geliştirilmesi

Abstract

In this study, it is aimed to determine the usability level of the original equation (JH_original) of the Jensen Haise method in the monthly average reference evapotranspiration (ETo) estimations in the Black Sea Region conditions and to calibrate it in accordance with the local climatic conditions. First of all, the climate classification of the region was made according to the De Martonne aridity index method. Then modified Jensen Haise (JH_modified) equations were developed for each climate class, using the Microsoft Excel program solver add-on. ETo calculations were made using monthly average temperature, relative humidity, wind speed and sunshine duration data (1927 – 2020) of the cities in the region. The ETo values estimated by the JH_original and JH_modified equations were compared with the values determined using the FAO 56 Penman Monteith (PM) equation. While the eastern part of the region is in the very humid climate class, the climate has changed to humid and semi humid towards the west along the coastline. It has been observed that the semi arid – humid climate is dominant in the interior parts, which receive less precipitation compared to the coastal part in region. The annual average ETo values obtained with FAO 56 PM for very humid, humid, semi humid and semi arid – humid conditions were determined as 1.779 mm day-1, 2.349 mm day-1, 2.280 mm day-1 and 2.631 mm day-1, respectively. For the same climate classes, the values estimated with JH_original were obtained as 2.146 mm day-1, 2.697 mm day-1, 2.566 mm day-1 and 2.562 mm day-1. The values estimated using JH_modified were determined as 1.775 mm day-1, 2.319 mm day-1, 2.238 mm day-1 and 2.629 mm day-1, respectively. The MAPE for the ETo values estimated using JH_original varied between %19.357 – 45.561 on the basis of city. Although the accuracy of ETo values is “acceptable” (MAPE= 20 – 50%), JH_original performed poorly in all climate class conditions. For this equation, the MAE, which changed between 0.428 – 0.596 mm day-1 on the basis of climate class, decreased by 62.14 – 92.76% after calibration and regressed to 0.031 – 0.205 mm day-1 range for JH_modified. JH_modified showed the best and worst performances in very humid (MAPE= 2.298%) and semi humid (MAPE= 13.631%) conditions, respectively. Its performance in humid and semi arid – humid conditions was close to each other (MAPE= 8.937 – 9.012). The accuracy of the ETo values estimated using this equation was determined as “good” (MAPE= 10 – 20%) in semi humid conditions and “very good” (MAPE= ≤ 10%) in other climate class conditions. It was concluded that the JH_modified equation is a good alternative and can be used to estimate ETo in humid conditions İn the absence of sufficient data for FAO 56 PM.