Digital Soil Mapping (DSM) Using a GIS-Based RF Machine Learning Model: The Case of Strandzha Mountains (Thrace Peninsula, Türkiye)

Küçük Resim Yok

Tarih

2024

Dergi Başlığı

Dergi ISSN

Cilt Başlığı

Yayıncı

Tech Science Press

Erişim Hakkı

info:eu-repo/semantics/closedAccess

Özet

This study assessed and mapped the spatial distribution of soil types and properties developed under the forest cover of the Strandzha Mountains of T & uuml;rkiye. The study was conducted on a micro-scale in the riparian zone of the Balaban River, which characterizes the soils distributed in the mountainous area. The effect of environmental factors on the spatial distribution of soil types and properties was also determined. To gather data, soil sampling, laboratory analysis, data processing and mapping were sequentially performed. These data were analyzed using the Geographical Information System (GIS) based Random Forest (RF) machine learning technique. Digital Soil Mapping (DSM) was developed with satisfactory performance. DSM suggests that the factors affecting the spatial distribution of soil types and properties in the sample area are, from most important to least important, topography (50.77%), climate (28.14%), organisms (8.22%), parent material (7.24%) and time (5.63%). With the contributions of all these factors in different proportions, it was determined that soils belonging to the Entisol and then Inceptisol orders were the most widespread in the sample area. The study results revealed that the GIS-based RF machinelearning technique can be used as a reliable tool for the development of DSM in mountainous terrains.

Açıklama

Anahtar Kelimeler

DSM, GIS, RF, soil, Strandzha Mountains

Kaynak

Revue Internationale De Geomatique

WoS Q Değeri

N/A

Scopus Q Değeri

Cilt

33

Sayı

Künye