Effect of vermicompost application on the development of plant properties and root architecture analysis with machine learning in <i>Buxus herlandii</i>


Sari O., Enginsu E., Çelikel F. G.

FOLIA HORTICULTURAE, 2025 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Basım Tarihi: 2025
  • Doi Numarası: 10.2478/fhort-2025-0005
  • Dergi Adı: FOLIA HORTICULTURAE
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Agricultural & Environmental Science Database, BIOSIS, CAB Abstracts, Food Science & Technology Abstracts, Directory of Open Access Journals
  • Ondokuz Mayıs Üniversitesi Adresli: Evet

Özet

The effects of liquid vermicompost (commercial product) on nutrient content, root architecture, and plant development were studied at doses of 0, 10, 20, 40, and 80 mL center dot pot(-1). Significant increases in plant height (3.5%), shoot length (25%), leaf width (16.9%), and leaf length (15.8%) were observed at the 40 mL center dot pot(-1) application compared to the control group. The highest number of shoots was observed at 10 mL center dot pot(-1), while the 80 mL center dot pot(-1) application led to a 3.9% reduction in shoot count. Root architecture showed a general decline compared to the control, though root length and tips number increased with 10 mL center dot pot(-1), and root volume was highest at 40 mL center dot pot(-1). However, high doses (40 and 80 mL center dot pot(-1)) caused a decrease in root surface area, forks number, and root crossings number. The highest nitrogen (31%) and manganese (57%) values were found at 10 mL center dot pot(-1). Phosphorus (-41%) and magnesium (40%) were lowest at 80 mL center dot pot(-1), while zinc (-46%) was lowest at 10 mL center dot pot(-1). The highest potassium content was recorded at 40 mL center dot pot(-1) (58%). The highest calcium (1.2%), iron (23%), and copper (77%) levels were obtained at 20 mL center dot pot(-1). Machine learning algorithms used for root growth prediction showed the following performance ranking: PART > J48 > Multilayer Perceptron > Multi-Class Classifier. These findings provide valuable insights for predicting root growth in Buxus crops