Estimation of mean monthly air temperatures in Turkey

Cobaner M., ÇITAKOĞLU H., Kisi O., HAKTANIR T.

COMPUTERS AND ELECTRONICS IN AGRICULTURE, cilt.109, ss.71-79, 2014 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 109
  • Basım Tarihi: 2014
  • Doi Numarası: 10.1016/j.compag.2014.09.007
  • Dergi Adı: COMPUTERS AND ELECTRONICS IN AGRICULTURE
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.71-79
  • Anahtar Kelimeler: ANN, ANFIS, Maximum monthly temperature, Minimum monthly temperature, Mean monthly temperature, ARTIFICIAL NEURAL-NETWORK, SOLAR-RADIATION
  • Erciyes Üniversitesi Adresli: Evet

Özet

One of the most important climatic factors influencing growth, development and yield of crops, which involve a countless number of biochemical reactions, is temperature. It is also one of the most effective explanatory variables of the evapotranspiration estimation models such as Hargreaves, Rich and Thornthwaite needed for irrigation scheduling policies. Using artificial neural networks (ANN), adaptive neuro-fuzzy inference system (ANFIS), and multiple linear regression (MLR) models, means of maximum, minimum, and average monthly temperatures are estimated as a function of geographical coordinates and month number for any location in Turkey. The monthly data measured by the General Directorate of State Meteorological Works at 275 stations having records of at least 20 years are used in developing the models. The latitude, longitude, and altitude of the location, and the month number are used as the input variables, and each of the mean monthly maximum, minimum, and average air temperatures is computed as the output variable. The observed values are compared versus those predicted by the ANN, ANFIS, and MLR models by evaluating their errors, and as a result the ANFIS model turns out to be the best. (C) 2014 Elsevier B.V. All rights reserved.