Using geostatistics and machine learning models to analyze the influence of soil nutrients and terrain attributes on lead prediction in forest soils

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F60460709%3A41210%2F24%3A97963" target="_blank" >RIV/60460709:41210/24:97963 - isvavai.cz</a>

Result on the web

<a href="https://link.springer.com/content/pdf/10.1007/s40808-023-01890-4.pdf" target="_blank" >https://link.springer.com/content/pdf/10.1007/s40808-023-01890-4.pdf</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/s40808-023-01890-4" target="_blank" >10.1007/s40808-023-01890-4</a>

Result language

angličtina

Original language name

Using geostatistics and machine learning models to analyze the influence of soil nutrients and terrain attributes on lead prediction in forest soils

Original language description

The study aimed at investigating the possibility of predicting lead (Pb) in forest soils by combining terrain attributes and soil nutrients using geostatistics and machine learning algorithms (MLAs). The study was partitioned into three categories: predicting Pb in forest soil using terrain attributes and RK (Context 1); predicting Pb in forest soil using soil nutrients and RK (Context 2); and lastly predicting Pb in forest soils using a combination of soil nutrients, terrain attributes, and RK (Context 3). Stochastic Gradient Boosting-regression kriging (SGB-RK), cubist regression kriging (CUB_RK), quantile regression forest kriging(QRF_RK) and k nearest neighbour regression kriging (KNN_RK) were the modeling approaches used in the estimation of lead (Pb) concentration in forest soil. The results showed that combining the terrain attribute as an auxiliary dataset with QRF_RK proved to be the most effective method for predicting Pb in forest soil (context 1). The most effective method for predicting Pb in forest soil was to combine soil nutrients as an auxiliary dataset with SGB_RK (context 2). Combining cubist_RK with an ancillary dataset of soil nutrients and terrain attributes is the most effective method for predicting Pb in forest soils (context 3). In addition, combining ancillary variables such as soil nutrients and terrain attributes with cubist_RK generated the best results for estimating Pb concentration in forest soils. It was found that applying a robust digital soil mapping (DSM) model in combination with terrain attributes and soil nutrients is efficient in predicting the spatial distribution and estimation of uncertainty levels of lead (Pb) in forest soils based on the model’s accuracy parameters.

Czech name

—

Czech description

—

Type

J_imp - Article in a specialist periodical, which is included in the Web of Science database

CEP classification

—

OECD FORD branch

40104 - Soil science

Project

Result was created during the realization of more than one project. More information in the Projects tab.

Continuities

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>S - Specificky vyzkum na vysokych skolach

Publication year

2024

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Name of the periodical

MODELING EARTH SYSTEMS AND ENVIRONMENT

ISSN

2363-6203

e-ISSN

2363-6203

Volume of the periodical

10

Issue of the periodical within the volume

2

Country of publishing house

NL - THE KINGDOM OF THE NETHERLANDS

Number of pages

14

Pages from-to

2099-2112

UT code for WoS article

001116446700002

EID of the result in the Scopus database

2-s2.0-85177770658