Assessing Effects of Agronomic Nitrogen Management on Crop Nitrogen Use and Nitrogen Losses in the Western Canadian Prairies

Abstract

Effective agronomic nitrogen management strategies ensure optimum productivity, reduce nitrogen losses, and enhance economic profitability and environmental quality. Farmers in western Canada make key decisions on formulation, rate, timing, and placement of fertilizer nitrogen that are suitable for soils, weather, and farming operations within which they operate. Suitability of agronomic nitrogen management options are assessed by estimates from linear interpolations and extrapolations of temporally and spatially discrete field-plot measurements of nitrogen responses. Such estimates do not account for non-linear and offsetting biogeochemical feedbacks of nitrogen cycles and cannot provide comprehensive nitrogen budgets for alternative nitrogen management options. These limitations can be overcome by using process-based agro-ecosystem models that adequately simulate basic processes of nitrogen biogeochemical cycles and are rigorously tested against site observations. Ecosys is a process-based ecosystem model that successfully simulated the biogeochemical feedbacks among nitrogen, carbon, and phosphorus cycles across different agro-ecosystems. This study deployed ecosys to generate spatially and temporally continuous estimates to assess crop nitrogen use and agronomic nitrogen losses from the crop fields across Alberta for alternative nitrogen fertilizer management scenarios. The study simulated effects of four nitrogen management scenarios: fall banded urea, fall banded ESN (Environmentally Smart Nitrogen), spring banded urea, and spring banded ESN on nitrogen recovery and losses from barley fields on mid-slope landforms. These simulations were done at township grids of ∼10 km × 10 km over 2011-2015 utilizing provincial soil and climate datasets. Modeled annual N 2 O, N 2 , and NH 3 emissions, and nitrogen losses in surface runoff and sub-surface discharge were lower by about 25, 30, 70, and 40%, respectively, with spring banding than in fall banding across Alberta. Modeled barley yields and grain nitrogen uptake were similar in spring and fall banding, indicating agro-economic and environmental sustainability advantage of spring banding in Alberta. These modeled estimates were consistent with estimates based on plot and laboratory research for Mezbahuddin et al. Modeling Prairie Agronomic N Management Alberta and similar prairie conditions. This study pioneered a methodology of processbased agroecosystem modeling, which is replicable and scalable to assess cumulative impacts of alternative agronomic nitrogen management options on crop production and the environment on provincial, regional, federal, continental, and global scales.

Knowledge-guided machine learning can improve carbon cycle quantification in agroecosystems

“…Although a large number of PB models exist for ecosystem carbon cycle modeling, models that incorporate sufficiently explicit representations of processes and are well-validated have more potential to benefit AI models, especially where no or few real-world samples are available to train the models. The PB model used in this study, ecosys, contains comprehensive first-principles descriptions of carbon transformation and translocation processes in plants and soil, and has been well-validated for different crop types and regions 13,[44][45][46][47] . It provides valuable basic knowledge to guide the structural design and training of the KGML model.…”

Combining Remotely Sensed Evapotranspiration and an Agroecosystem Model to Estimate Center‐Pivot Irrigation Water Use at High Spatio‐Temporal Resolution

“…Ecosys, an advanced agroecosystem model based on biophysical and biochemical mechanisms, uses the multi-layered soil-root-canopy system to track the water, energy, carbon, and nutrient cycles (Grant, 1995(Grant, , 1997Grant et al, 1993). Ecosys can simulate major agricultural management practices, including irrigation (Grant et al, , 2004, fertilizer (Grant, Juma, Robertson, Izaurralde, & McGill, 2001), crop rotation (Grant, 1997), and tillage (Grant, 1997), which has been extensively validated in many agricultural ecosystems (Grant, 1995;Grant & Flanagan, 2007;Grant et al, , 2011Grant et al, , 1993Grant et al, , 1999Mezbahuddin et al, 2020;…”

Quantifying carbon budget, crop yields and their responses to environmental variability using the ecosys model for U.S. Midwestern agroecosystems

“…Ecosys is an advanced mechanistic ecosystem model developed to simulate water, energy, carbon, and nutrient cycles simultaneously for various ecosystems, including agroecosystems at the hourly step (Figure 1a) (Grant, 2001). It is one of the very few models that are formulated primarily based on biophysical and biochemical principles, with fully connected balances and interactions for water, energy, carbon and nutrient cycles in the soil-plant-atmosphere continuum, and has been extensively validated in various ecosystems ranging from agricultural (Grant et al, , 2011Mezbahuddin et al, 2020) to forest systems (Grant et al, 2010.…”