Publication: Patlıcan Bitkisinin Sulama Programlamasının Belirlenmesinde Bulanık Mantık Uygulamaları
Loading...
Date
Authors
Journal Title
Journal ISSN
Volume Title
Abstract
Bu çalışma Samsun ili Bafra ilçesinde 2015 ve 2016 yıllarında arazi ve saksı olmak üzere iki farklı koşulda yetiştirilen patlıcan bitkisi sulama programlamasının belirlenmesi için yürütülmüştür. Tesadüf blokları deneme desenine göre üç tekrarlamalı olarak planlanan çalışma tam su konusu (S1:%100-kontrol) ve bitkilerde farklı düzeylerde su stresi meydana getirecek dört kısıntılı (S2: %75, S3: %50, S4:%25, S5: %0) olmak üzere 5 konudan oluşmaktadır. Araştırmanın ilk aşamasında arazi çalışmaları yapılmış ikinci aşamasında ise araziden elde edilen bitki, toprak ve iklim verileri kullanılarak bulanık mantık yöntemine göre modellemeleri yapılmıştır. Buna göre araştırma süresi boyunca S1, S2, S3, S4 ve S5 konularına iki yıl ortalaması olarak sırasıyla 400, 313, 231, 146 ve 63 mm sulama suyu uygulanmış ve bahsedilen bu konulara ait bitki su tüketim değerleri sırasıyla 584, 480, 377, 300 ve 246 mm olarak kaydedilmiştir. Araştırmada Bafra koşullarında patlıcan bitkisi için 0-60 cm toprak derinliğindeki elverişli nemin %30'u tüketildiğinde tarla kapasitesi düzeyine ulaştıracak miktardaki sulama suyu uygulamasının mevsim boyunca en iyi verimi elde edilebileceği, mevsim boyunca bitki su stres indeksi (CWSI) değerinin 0.20 civarında sürdürülmesi gerektiği, mevsimlik verim tepki etmeni (ky) değerinin 1.25 olarak elde edildiği böylelikle bitkinin su stresine karşı duyarlı olduğu ve uygulanan tekniklerin yörede patlıcan bitkisi sulama programlaması amacıyla kullanılabileceği sonuçlarına ulaşılmıştır. Elde edilen sonuçlar ile bulanık mantık yöntemine göre Sugeno ve Mamdani çıkarım mekanizmasında bitki su tüketimi, bitki katsayısı, bitki su stres indeksi ve verim değerlerinin modellemesi yapılmıştır. Bahsedilen bu yöntemler içerisinde Sugeno yöntemi birçok model de daha başarılı sonuçlar vermiştir.
This study was conducted to determine irrigation scheduling for eggplant grown in two different conditions as land and pot in Bafra district of Samsun/Turkey during period of 2015-2016. During the study randomized blocks experimental design were applied to monitor water-stress in three repetitive sequences to the five plant subjects. First subject of researches, S1, were chosen as control group, and supplied 100% water in the irrigation process, then other four subjects of plants were irrigated by decreasing the supplied water amount 25% in total of five different subjects represented as S1:100%, S2:75%, S3:50%, S4:25%, S5:0%. The research was divided two different sections. In the first section, land works were performed, and data collected for soil and climate conditions. In the second section, modelling of collected data combinations were performed by applying fuzzy logic. In the research, five group of plants, S1, S2, S3, S4, and S5 were given respectively 400 mm, 313 mm, 231 mm, 146 mm and 63 mm irrigation water as the average of two years, and each group of plants consumed respectively 584 mm, 480 mm, 377 mm, 300 mm and 246 mm water. In the studied region, the results released that applied techniques were applicable for plant irrigation scheduling programs. When the results were evaluated, it was found that the best yield for eggplant can be accessible throughout the season when irrigation was carried in the field capacity in the case of having moisture level lower than the usable moisture in the soil. Additionally, the following outcomes were obtained: Eggplant can be used for the irrigation process with seasonal yield reaction factor value of 1.25 (ky), but crop water-stress index is required to be sustain around 0.20 throughout the season because the plant is sensitive to the water-stress. By evaluating all collected outcomes with inference mechanisms of Sugeno and Mamdani in fuzzy logic method, the modelling of evapotranspiration, crop coefficient, crop water-stress index and yield values were carried out. Among these methods, the Sugeno method has been more successful in many models.
This study was conducted to determine irrigation scheduling for eggplant grown in two different conditions as land and pot in Bafra district of Samsun/Turkey during period of 2015-2016. During the study randomized blocks experimental design were applied to monitor water-stress in three repetitive sequences to the five plant subjects. First subject of researches, S1, were chosen as control group, and supplied 100% water in the irrigation process, then other four subjects of plants were irrigated by decreasing the supplied water amount 25% in total of five different subjects represented as S1:100%, S2:75%, S3:50%, S4:25%, S5:0%. The research was divided two different sections. In the first section, land works were performed, and data collected for soil and climate conditions. In the second section, modelling of collected data combinations were performed by applying fuzzy logic. In the research, five group of plants, S1, S2, S3, S4, and S5 were given respectively 400 mm, 313 mm, 231 mm, 146 mm and 63 mm irrigation water as the average of two years, and each group of plants consumed respectively 584 mm, 480 mm, 377 mm, 300 mm and 246 mm water. In the studied region, the results released that applied techniques were applicable for plant irrigation scheduling programs. When the results were evaluated, it was found that the best yield for eggplant can be accessible throughout the season when irrigation was carried in the field capacity in the case of having moisture level lower than the usable moisture in the soil. Additionally, the following outcomes were obtained: Eggplant can be used for the irrigation process with seasonal yield reaction factor value of 1.25 (ky), but crop water-stress index is required to be sustain around 0.20 throughout the season because the plant is sensitive to the water-stress. By evaluating all collected outcomes with inference mechanisms of Sugeno and Mamdani in fuzzy logic method, the modelling of evapotranspiration, crop coefficient, crop water-stress index and yield values were carried out. Among these methods, the Sugeno method has been more successful in many models.
Description
Tez (doktora) -- Ondokuz Mayıs Üniversitesi, 2018
Libra Kayıt No: 125393
Libra Kayıt No: 125393
Citation
WoS Q
Scopus Q
Source
Volume
Issue
Start Page
End Page
272
