100 m soil landscape grids of Canada

Area of Science:

• Soil Science
• Geospatial Analysis
• Environmental Modeling

Background:

• Accurate soil data is crucial for environmental assessments and land management.
• Previous soil mapping efforts may lack the resolution or comprehensiveness required for detailed analysis.
• The development of high-resolution digital soil information is a global priority.

Purpose of the Study:

• To create high-resolution (100m) digital grids of soil type and key soil attributes for Canada.
• To provide a foundational dataset for national and regional soil organic carbon stock inventories.
• To enable robust carbon sequestration assessments and support various environmental modeling applications.

Main Methods:

• Utilized the latest soil point data and available co-variables.
• Employed machine learning algorithms for grid development.
• Conducted model-based statistical validations and limited independent statistical validations.

Main Results:

• Successfully developed 100m resolution grids of soil type and selected soil attributes across Canada.
• Machine learning model-based validations yielded satisfactory results.
• Independent statistical validations, though limited, were also performed.

Conclusions:

• The 100m soil landscape grids offer a valuable resource for national soil organic carbon stock inventory and carbon sequestration assessment.
• These grids have significant potential for applications in soil health, erosion modeling, land suitability analysis, and other large-area modeling.
• The high-resolution soil data supports improved environmental management and policy decisions in Canada.