AI-Enhanced Projections of Hazelnut Production Under Climate Change: a Regional Analysis from Turkey
Küçük Resim Yok
Tarih
2025
Yazarlar
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
Springer
Erişim Hakkı
info:eu-repo/semantics/closedAccess
Özet
Hazelnut (Corylus avellana L.) cultivation, a cornerstone of Turkey's fruit production and exports, faces mounting challenges due to climate change. This study analyzes the effects of climatic variables and soil properties on hazelnut yield in two key Turkish regions-Eastern Black Sea and D & uuml;zce-by combining 20 years of agronomic data with machine learning techniques. Yield predictions under different Shared Socioeconomic Pathway (SSP) climate scenarios were generated using Random Forest and Gradient Boosting models, achieving high predictive performance (R-2 > 0.90). Results show that low-emission scenarios (SSP1-1.9) foster favorable growing conditions, while high-emission scenarios (SSP5-8.5) lead to significant yield reductions (up to 17%), mainly due to increased humidity-induced fungal risk and frost events. Soil properties such as organic matter content and water-holding capacity also played critical roles in determining yield variability. The study highlights the potential of AI-driven climate modeling to guide adaptive cultivation practices, such as frost-resistant cultivars and humidity control strategies, ensuring sustainable hazelnut production in the face of future climate uncertainties.
Açıklama
Anahtar Kelimeler
Hazelnut yield, Climate change, SSP scenarios, Machine learning, Regional agriculture, Agricultural sustainability
Kaynak
Applied Fruit Science
WoS Q Değeri
N/A
Scopus Q Değeri
N/A
Cilt
67
Sayı
5
