{"id":{"repo_id":"abertay","oai_identifier":"oai:rke.abertay.ac.uk:studenttheses/24eb5811-3be1-44b0-8c53-4688590850cc"},"canonical_url":"https://search.dev.ndltd.org/etd/abertay/oai:rke.abertay.ac.uk:studenttheses/24eb5811-3be1-44b0-8c53-4688590850cc","repository":{"repo_id":"abertay","name":"Abertay University","base_url":"https://rke.abertay.ac.uk/ws/oai"},"display":{"title":"Predicting soil properties using machine learning: a smartphone solution for stakeholders","abstract":"Soil health is foundational to agricultural productivity, ecosystem sustainability, and climate regulation, yet traditional methods for assessing soil properties are labour-intensive, costly, and limited in spatial resolution. This thesis addresses these limitations by integrating machine learning and Digital Soil Mapping (DSM) with smartphone-based imagery to enhance the accuracy, accessibility, and affordability of soil property assessment. Using strategically sampled field data from diverse Scottish environments, detailed laboratory analysis established a robust baseline dataset for key soil attributes, including pH, organic carbon, moisture content, bulk density, and texture.<br/><br/>Advanced machine learning algorithms were developed and optimised to predict soil properties across Scotland, significantly increasing spatial resolution and capturing micro-scale variability. Incorporation of smartphone-captured soil images analysed through Convolutional Neural Networks (CNNs) further improved prediction accuracy, demonstrating the substantial value of visual soil characteristics in modelling efforts. The suitability of CNN-derived features was rigorously tested against traditional features, both with and without calibration, confirming that CNN features provided superior performance in predictive accuracy. The resulting high-resolution soil property maps, with quantified uncertainty intervals, represent a transformative improvement over conventional soil maps, enabling targeted agricultural and conservation strategies.<br/><br/>A user-friendly smartphone application was created to operationalise these advancements, empowering farmers, policymakers, and land managers with instant, site-specific soil information. Users can obtain site-specific soil property predictions by simply capturing a geotagged soil image. In cross validated tests, our best models achieved R² of 0.64 for bulk density, 0.61 for moisture content, 0.39 for pH, and 0.50 for LOI. <br/><br/>This research advances the science of digital soil mapping and directly contributes to sustainable land management practices, precision agriculture, and environmental conservation, offering a replicable and scalable framework suitable for global adoption in diverse geographical settings.","abstract_html":"Soil health is foundational to agricultural productivity, ecosystem sustainability, and climate regulation, yet traditional methods for assessing soil properties are labour-intensive, costly, and limited in spatial resolution. This thesis addresses these limitations by integrating machine learning and Digital Soil Mapping (DSM) with smartphone-based imagery to enhance the accuracy, accessibility, and affordability of soil property assessment. Using strategically sampled field data from diverse Scottish environments, detailed laboratory analysis established a robust baseline dataset for key soil attributes, including pH, organic carbon, moisture content, bulk density, and texture.&lt;br/&gt;&lt;br/&gt;Advanced machine learning algorithms were developed and optimised to predict soil properties across Scotland, significantly increasing spatial resolution and capturing micro-scale variability. Incorporation of smartphone-captured soil images analysed through Convolutional Neural Networks (CNNs) further improved prediction accuracy, demonstrating the substantial value of visual soil characteristics in modelling efforts. The suitability of CNN-derived features was rigorously tested against traditional features, both with and without calibration, confirming that CNN features provided superior performance in predictive accuracy. The resulting high-resolution soil property maps, with quantified uncertainty intervals, represent a transformative improvement over conventional soil maps, enabling targeted agricultural and conservation strategies.&lt;br/&gt;&lt;br/&gt;A user-friendly smartphone application was created to operationalise these advancements, empowering farmers, policymakers, and land managers with instant, site-specific soil information. Users can obtain site-specific soil property predictions by simply capturing a geotagged soil image. In cross validated tests, our best models achieved R² of 0.64 for bulk density, 0.61 for moisture content, 0.39 for pH, and 0.50 for LOI. &lt;br/&gt;&lt;br/&gt;This research advances the science of digital soil mapping and directly contributes to sustainable land management practices, precision agriculture, and environmental conservation, offering a replicable and scalable framework suitable for global adoption in diverse geographical settings.","abstract_has_math":false,"creators":["Khan, Adnan"],"institution":"Abertay University","degree_name":"PhD","degree_level":"Doctoral Thesis","degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":["Jorat, Mohammadehsan","Howson, Thomas"],"committee_chairs":[],"committee_members":[],"year":2025,"date_issued":"2025-11-3","date_published":"2025-11-3","updated_at":"2026-07-24T00:50:29Z","subjects":["Digital soil mapping","Convolutional neural network","Machine learning","Smartphone","Soil health"],"languages":["eng"],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["oai:rke.abertay.ac.uk:studenttheses/24eb5811-3be1-44b0-8c53-4688590850cc"],"render_values":[{"text":"oai:rke.abertay.ac.uk:studenttheses/24eb5811-3be1-44b0-8c53-4688590850cc","href":null,"code":true}]}]},"links":{"outbound_url":"https://rke.abertay.ac.uk/en/studentTheses/24eb5811-3be1-44b0-8c53-4688590850cc","outbound_label":"Repository record","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Jorat, Mohammadehsan","Howson, Thomas"]},{"key":"dc:creator","label":"Author","values":["Khan, Adnan"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2025-11-3"]},{"key":"dc:date.issued","label":"Date","values":["2025-11-3"]},{"key":"dc:publisher.department","label":"Dc Publisher Department","values":["Faculty of Social and Applied Sciences"]},{"key":"dc:publisher.institution","label":"Dc Publisher Institution","values":["Abertay University"]},{"key":"dc:relation.isreferencedby","label":"Dc Relation Isreferencedby","values":["https://rke.abertay.ac.uk/en/studentTheses/24eb5811-3be1-44b0-8c53-4688590850cc"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"dc:type.qualificationlevel","label":"Dc Type Qualificationlevel","values":["Doctoral Thesis"]},{"key":"dc:type.qualificationname","label":"Dc Type Qualificationname","values":["PhD"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Digital soil mapping","Convolutional neural network","Machine learning","Smartphone","Soil health"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["eng"]},{"key":"dc:rights.embargodate","label":"Dc Rights Embargodate","values":["2027-05-31"]},{"key":"dc:rights.embargoreason","label":"Dc Rights Embargoreason","values":["/dk/atira/pure/core/document/studentthesisembargoreason/publicationissues"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["oai:rke.abertay.ac.uk:studenttheses/24eb5811-3be1-44b0-8c53-4688590850cc","https://rke.abertay.ac.uk/en/studentTheses/24eb5811-3be1-44b0-8c53-4688590850cc"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Soil health is foundational to agricultural productivity, ecosystem sustainability, and climate regulation, yet traditional methods for assessing soil properties are labour-intensive, costly, and limited in spatial resolution. This thesis addresses these limitations by integrating machine learning and Digital Soil Mapping (DSM) with smartphone-based imagery to enhance the accuracy, accessibility, and affordability of soil property assessment. Using strategically sampled field data from diverse Scottish environments, detailed laboratory analysis established a robust baseline dataset for key soil attributes, including pH, organic carbon, moisture content, bulk density, and texture.<br/><br/>Advanced machine learning algorithms were developed and optimised to predict soil properties across Scotland, significantly increasing spatial resolution and capturing micro-scale variability. Incorporation of smartphone-captured soil images analysed through Convolutional Neural Networks (CNNs) further improved prediction accuracy, demonstrating the substantial value of visual soil characteristics in modelling efforts. The suitability of CNN-derived features was rigorously tested against traditional features, both with and without calibration, confirming that CNN features provided superior performance in predictive accuracy. The resulting high-resolution soil property maps, with quantified uncertainty intervals, represent a transformative improvement over conventional soil maps, enabling targeted agricultural and conservation strategies.<br/><br/>A user-friendly smartphone application was created to operationalise these advancements, empowering farmers, policymakers, and land managers with instant, site-specific soil information. Users can obtain site-specific soil property predictions by simply capturing a geotagged soil image. In cross validated tests, our best models achieved R² of 0.64 for bulk density, 0.61 for moisture content, 0.39 for pH, and 0.50 for LOI. <br/><br/>This research advances the science of digital soil mapping and directly contributes to sustainable land management practices, precision agriculture, and environmental conservation, offering a replicable and scalable framework suitable for global adoption in diverse geographical settings."]},{"key":"dc:title","label":"Title","values":["Predicting soil properties using machine learning: a smartphone solution for stakeholders"]}]}],"canonical_facts":{"dc:contributor.advisor":["Jorat, Mohammadehsan","Howson, Thomas"],"dc:creator":["Khan, Adnan"],"dc:date":["2025-11-3"],"dc:date.issued":["2025-11-3"],"dc:description.abstract":["Soil health is foundational to agricultural productivity, ecosystem sustainability, and climate regulation, yet traditional methods for assessing soil properties are labour-intensive, costly, and limited in spatial resolution. This thesis addresses these limitations by integrating machine learning and Digital Soil Mapping (DSM) with smartphone-based imagery to enhance the accuracy, accessibility, and affordability of soil property assessment. Using strategically sampled field data from diverse Scottish environments, detailed laboratory analysis established a robust baseline dataset for key soil attributes, including pH, organic carbon, moisture content, bulk density, and texture.<br/><br/>Advanced machine learning algorithms were developed and optimised to predict soil properties across Scotland, significantly increasing spatial resolution and capturing micro-scale variability. Incorporation of smartphone-captured soil images analysed through Convolutional Neural Networks (CNNs) further improved prediction accuracy, demonstrating the substantial value of visual soil characteristics in modelling efforts. The suitability of CNN-derived features was rigorously tested against traditional features, both with and without calibration, confirming that CNN features provided superior performance in predictive accuracy. The resulting high-resolution soil property maps, with quantified uncertainty intervals, represent a transformative improvement over conventional soil maps, enabling targeted agricultural and conservation strategies.<br/><br/>A user-friendly smartphone application was created to operationalise these advancements, empowering farmers, policymakers, and land managers with instant, site-specific soil information. Users can obtain site-specific soil property predictions by simply capturing a geotagged soil image. In cross validated tests, our best models achieved R² of 0.64 for bulk density, 0.61 for moisture content, 0.39 for pH, and 0.50 for LOI. <br/><br/>This research advances the science of digital soil mapping and directly contributes to sustainable land management practices, precision agriculture, and environmental conservation, offering a replicable and scalable framework suitable for global adoption in diverse geographical settings."],"dc:identifier":["oai:rke.abertay.ac.uk:studenttheses/24eb5811-3be1-44b0-8c53-4688590850cc","https://rke.abertay.ac.uk/en/studentTheses/24eb5811-3be1-44b0-8c53-4688590850cc"],"dc:language":["eng"],"dc:publisher.department":["Faculty of Social and Applied Sciences"],"dc:publisher.institution":["Abertay University"],"dc:relation.isreferencedby":["https://rke.abertay.ac.uk/en/studentTheses/24eb5811-3be1-44b0-8c53-4688590850cc"],"dc:rights.embargodate":["2027-05-31"],"dc:rights.embargoreason":["/dk/atira/pure/core/document/studentthesisembargoreason/publicationissues"],"dc:subject":["Digital soil mapping","Convolutional neural network","Machine learning","Smartphone","Soil health"],"dc:title":["Predicting soil properties using machine learning: a smartphone solution for stakeholders"],"dc:type":["Thesis"],"dc:type.qualificationlevel":["Doctoral Thesis"],"dc:type.qualificationname":["PhD"]},"updated_at":"2026-07-24T00:50:29Z"}