Predicting Future Winter Subsurface Drainage Dynamics for Subsurface-Drained Croplands in Cold Climates under Climate

Subsurface drainage is a commonly practiced agricultural management practice in North America to improve crop yield by removing excess water from the field. However, previous studies also discovered that the nutrient leached from the subsurface drainage was a leading contributor to the nutrient loss to surface water bodies. Recent in-situ experiments in subsurface-drained croplands in Eastern Canada revealed that winter is critical in determining the region's annual subsurface drainage flow and associated nutrient loss. However, it has been observed that the winter meteorological conditions are continuously changing in Canada. This thesis investigates the impact of warmer winter on winter subsurface drainage dynamics in Eastern Canada's subsurface-drained croplands by using one of the most detailly described process-based bio-physical models, namely the Root Zone Water Quality-Simultaneous Heat and Water (RZ-SHAW) model. Through a multifaceted research approach, the study explores soil freezing dynamics for croplands in Canada under warmer winters, optimizes the RZ-SHAW model simulation time via a custom parallel-distributed computing framework (RS-DPCF), evaluates the model's winter subsurface drainage simulation accuracy, and predicts future winter subsurface drainage dynamics under climate change scenarios.Key findings reveal the nuanced response of soil freezing to warmer winters, highlighting an increase in soil frozen depth with rising temperatures in specific scenarios where the energy gained from reduced snow insulation outweighs the energy gained through increasing air temperature. This dynamic underscore the critical balance between snow cover reduction and soil energy balance alterations due to climate change. The development of the parallel-distributed RS-DPCF significantly improved model efficiency, enabling faster, scalable calibrations and simulations across multiple sites. The RZ-SHAW model was calibrated and validated and was evaluated to deliver satisfactory performance in simulating winter subsurface drainage for two croplands in Eastern Canada. Comparative analyses of the RZ-SHAW model with machine learning models identified the Cubist and SVM-RBF as efficient alternatives for short-term simulations. However, long-term projections underscored the challenge of temporal limited and unbalanced winter subsurface drainage data in capturing winter hydrology’s full complexity for subsurface-drained croplands in cold climates.Future projections indicate a substantial increase in winter subsurface drainage volume and frequency, with a shift towards a more evenly distributed drainage pattern closely aligning with the monthly precipitation pattern. These shifts, driven by the simulated shorter snow coverage, advanced snowmelt timing, and reduced soil freezing periods, suggest February will emerge as a peak drainage month, reversing traditional patterns and highlighting the evolving challenge of managing winter subsurface drainage under climate change.This research contributes to the broader understanding of agricultural hydrology's response to climate change during winter, offering valuable insights for developing adaptive management strategies in cold climate regions

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.