Enhancing Soil Texture and Bulk Density Mapping Using Soil Grids and Machine Learning - A Comparative Analysis with Observed Data

Aug 5, 2026

Upload by Rowan

Digital soil mapping supports reliable characterization of soil variability and strengthens sustainable land management decisions. This study evaluates the accuracy of SoilGrids for the Kurdistan Region of Iraq by comparing global predictions with ground truth sampling data. Clay, silt, and sand fractions and bulk density are assessed through comparative error analysis, showing significant differences between observed values and SoilGrids estimates. Machine learning, especially Extreme Gradient Boosting (XGBoost), improves estimation accuracy, reducing mean errors and improving RMSE and MAPE across soil fractions and bulk density.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages19

Reading Time48 min

AccessFree to download

Abstract

Introduction

Materials and Methods

Soil sampling and ground truth data

SoilGrids predictions

Machine learning models and evaluation

Results

Soil fractions comparison

Bulk density comparison

Discussion

Conclusion and Future Work

Research outputs 2022 to 2026 10-29-2024 Enhancing soil texture and bulk density mapping using soil gridsand machine learning: A comparative analysis with observed data Aram Ali Ismael O. Ismael Hewa T. Mustafa Diman Krwanji Edith Cowan University Akram O. Esmail Follow this and additional works at: [https://ro.ecu.edu.au/ecuworks2022-2026](https://ro.ecu.edu.au/ecuworks2022-2026) Part of the Natural Resources and Conservation Commons 10.9734/asrj/2024/v8i4163 Ali, A., Ismael, I. O., Mustafa, H. T., Krwanji, D., & Esmail, A. O. (2024) . Enhancing soil texture and bulk density mapping using soil grids and machine learning: A comparative analysis with observed data. Asian Soil Research Journal, 8(4), 61–78 . [https://doi.org/10.9734/asrj/2024/v8i4163](https://doi.org/10.9734/asrj/2024/v8i4163) [This Journal Article is posted at Research Online.](This Journal Article is posted at Research Online.)[ ](This Journal Article is posted at Research Online.)[https://ro.ecu.edu.au/ecuworks2022-2026/5587](https://ro.ecu.edu.au/ecuworks2022-2026/5587) Asian Soil Research Journal Volume 8, Issue 4, Page 61-78, 2024; Article no.ASRJ.125367 ISSN: 2582-3973 Enhancing Soil Texture and Bulk Density Mapping Using Soil Grids and Machine Learning: A Comparative Analysis with Observed Data Aram Ali a,b* , Ismael O. Ismael a, Hewa T. Mustafa a, Diman Krwanji c,d and Akram O. Esmail a a Soil and Water Department, College of Agricultural Engineering Sciences, Salahaddin UniversityErbil, Erbil, Kurdistan Region, Iraq. b University of Southern Queensland, Centre for Sustainable Agricultural Systems, West St, Toowoomba, QLD 4350, Australia. c Plant Protection Department, College of Agricultural Engineering Sciences, Salahaddin UniversityErbil, Erbil, Kurdistan Region, Iraq. d Edith Cowan University, 270 Joondalup Drive, Joondalup, WA 6027, Australia. Authors’ contributions This work was carried out in collaboration among all authors. All authors read and approved the final manuscript. Article Information DOI: [https://doi.org/10.9734/asrj/2024/v8i4163](https://doi.org/10.9734/asrj/2024/v8i4163) Open Peer Review History: This journal follows the Advanced Open Peer Review policy. Identity of the Reviewers, Editor(s) and additional Reviewers, peer review comments, different versions of the manuscript, comments of the editors, etc are available here: [https://www.sdiarticle5.com/review-history/125367](https://www.sdiarticle5.com/review-history/125367) Original Research Article Received: 20/08/2024 Accepted: 22/10/2024 Published: 29/10/2024 ABSTRACT Digital soil mapping plays a crucial role in understanding soil variability and informing sustainable land management practices. This study focuses on the Kurdistan Region of Iraq (KRI), evaluating the accuracy of SoilGrids, a global-scale soil mapping initiative, and exploring the efficacy of machine learning algorithms in refining soil properties estimations. The aim of this research was to assess and represent the physical parameters of soils effectively by comparing ground truth soil sampling data with data obtained from SoilGrids regarding clay, silt, and sand fractions and bulk density. Comparative analyses were conducted between ground truth soil sampling data and SoilGrids predictions, revealing significant differences across soil mineral fractions including clay, silt, sand fractions, and bulk density. The results showed that the mean clay fraction in the ground truth dataset differed notably from SoilGrids estimation, with a Mean Absolute Deviation (MAD) of 124.0 g kg-1 and Root Mean Square Error (RMSE) of 152.5. However, the integration of machine learning algorithms, particularly the Extreme Gradient Boosting (XG Boost) algorithm, showed promising results in improving accuracy. The XG Boost algorithm exhibited a relatively low MAD of 97.9 g kg-1 for clay fractions, indicating a better approximation of observed values compared to SoilGr

FAQs

What does the study evaluate in the Kurdistan Region of Iraq?

The study evaluates how accurately SoilGrids estimates key soil properties—clay, silt, sand fractions, and bulk density—using comparisons with ground truth soil sampling data.

Which machine learning algorithm shows the most promising results?

Extreme Gradient Boosting (XGBoost) shows promising performance by improving approximation of observed values compared with SoilGrids alone.

What improvements are reported when using machine learning for soil mapping?

Significant percentage improvements are reported in RMSE and MAPE across multiple soil fractions and bulk density, ranging from roughly 15% for clay to about 35% for sand fractions and about 20% for bulk density.

Aug 5, 2026

Upload by Rowan