Towards an improved methodology for modelling climate change impacts on cropping systems in cool climates

Area of Science:

• Agricultural Science
• Climate Change Modeling
• Soil Science

Background:

• Assessing climate change impacts on agricultural sustainability demands integrated models of soil-plant-atmospheric processes and dynamic farm management.
• Simplified modeling approaches often omit crucial feedback mechanisms, potentially skewing outcomes.

Purpose of the Study:

• To quantify the impact of simplified modeling approaches on crop yields, soil organic carbon (SOC) change, and nitrogen (N) losses under future climate scenarios in Canada.
• To compare simplified modeling methods against a comprehensive reference approach.

Main Methods:

• Utilized the DNDCv.CAN biogeochemical model at three Canadian locations.
• Compared simplified approaches (temperature/precipitation only, annual soil re-initialization, fixed fertilizer, fixed planting dates) with a comprehensive approach (detailed climate drivers, dynamic management, continuous budgets).
• Investigated impacts on spring wheat yields, SOC, N leaching, and N runoff across four time periods (1981-2100).

Main Results:

• Simplified approaches, particularly fixed fertilizer and soil re-initialization, led to substantial reductions in spring wheat yield estimates (up to 40%) in semi-arid regions for the 2071-2100 period.
• Soil re-initialization significantly altered SOC levels, N leaching, and N runoff in sub-humid locations from 2011 to 2100.
• Simplified methods generally impacted SOC levels across locations, highlighting the importance of continuous soil carbon dynamics.

Conclusions:

• Simplified modeling approaches inadequately capture critical feedback loops between climate, soil, crops, and management, compromising accurate assessments of agricultural systems under climate change.
• Modelers should enhance capabilities for simulating agronomic changes and utilize robust soil-plant-atmospheric process models for reliable future climate impact studies.