Yazar "Sarsici, N." seçeneğine göre listele

Listeleniyor 1 - 1 / 1
  • Küçük Resim Yok
    Öğe
    Using Google Earth Engine Machine Learning Algorithms, Soil Variable Effects on Soil Organic Carbon in Karabük Province/Turkiye: Using Google Earth Engine Machine Learning Algorithms
    (King Saud University, 2024) Coskun, M.; Coskun, S.; Dündar, Ö.; Sarsici, N.
    The study area is Karabük province, and the research topic is to examine the influence of soil-related variables on soil organic carbon in Karabük province. The aim of the study is to determine the relationship between digital soil mapping and the correlation analysis of soil variables that affect the carbon stock stored by the soil. In the study, data from SoilGrids was gathered using Google Earth Engine (GEE) machine learning methods. The JavaScript coding language was used to generate maps of SoilGrids data in GEE. These spatial data were processed using Geographic Information Systems software, and multiple linear regression analysis was performed using the “IBM SPSS 20.0? program. Clay, sand, silt, pH (in water), organic carbon density, mass density, coarse fractions, cation exchange capacity (CEC), and nitrogen were considered as soil variables. According to the results obtained, the pH of the surface soils (0–5 cm) of the study area was 58–7: clay g/kg; 104–400 g/kg; sand; 214–460; silt; 331–510 g/kg; organic carbon density: 380–562 dg/dm3; nitrogen density: 2 920–7 683 cg/kg; mass density: 93.00–136.00 g/kg; coarse particles: 55–239 (Per10000); CEC: 215–348 mmol/kg; and SOC values varied between 286–374 dg/kg. Soil organic carbon (SOC) stock amounts varied between 286 and 374 dg/kg in surface (0–5 cm) soils. As a consequence of the studies, it was revealed that nitrogen had the strongest link with SOC, whereas clay had the lowest relationship. © 2024 The Authors