Nitrogen and phosphorus control carbon sequestration in agricultural ecosystems: modelling carbon, nitrogen, and phosphorus balances at the Breton Plots with ecosys under historical and future climates

Abstract

Perennial legumes in crop rotations increase soil C sequestration from greater productivity with N2 fixation. Here, we corroborated increases in soil organic carbon (SOC) and harvests modelled in 5 yr wheat–oats–barley–alfalfa/brome–alfalfa/brome (5Y) vs. 2 yr wheat–fallow (WF) rotations with those measured from 1929 to 2018. Harvest and SOC gains of 100–150 g C m−2 yr−1 and 15–25 g C m−2 yr−1 were modelled and measured in 5Y vs. WF rotations with different fertilizer and manure amendments. Modelled gains were closely related to annualized rates of N2 fixation by alfalfa of 8–10 g N m−2 yr−1. However, N2 fixation also drove increases in modelled N2O emissions of ca. 0.06 g N m−2 yr−1, which partially offset gains in SOC. Gains in harvest, SOC, and N2O emissions of 60–90 g C m−2 yr−1, 15 g C m−2 yr−1, and 0.05 g N m−2 yr−1 were modelled and measured in both rotations with amendments of N + P relative to unamended treatments. Harvest and SOC gains were smaller, and leaching and N2O losses larger, with amendments of N without P. After 100 yr of RCP 8.5 climate change, harvests in WF changed little from those in baseline runs, whereas those in 5Y rose with N + P because of increased N2 fixation. SOC declined in WF with all amendments and could only be raised in 5Y with N + P amendments. These model findings indicated the importance of N2 fixation and P amendments in determining responses of agroecosystem productivity and C sequestration to climate change.

Machine learning models inaccurately predict current and future high-latitude C balances

“…Since increases in NPP generate increases in plant litter, underestimation of litter C follows from the NPP underestimation by the ML model. Root and shoot litter is highly labile, so increases in litter C typically lead to increases in microbial respiration in field studies (Adamczyk et al 2020) and in the ecosys model (Grant et al 2020). Indeed, we find that litter C is the leading driver of ML underestimation of 21st century R h , with increases in ML R h bias strongly tied to increases in litter C in 40% of grid cells (figure 7).…”

Assessing Different Plant‐Centric Water Stress Metrics for Irrigation Efficacy Using Soil‐Plant‐Atmosphere‐Continuum Simulation

“…Seed number and kernel mass are set during pre‐ and post‐anthesis growth stages to determine the yield upon harvest (Grant et al., 2011). More details about the biophysical and biochemical processes in ecosys can be found in the supplement of Grant, Lin, and Hernandez‐Ramirez (2020).…”

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

“…The physiological adaptation in the barley PFT was represented by raising the fractions of leaf proteins in rubisco and in mesophyll chlorophyll by 40%, which would simulate higher productivity and hence more rapid accumulation of grain biomasses in barley than in the wheat PFT in Grant et al (2011). These adjustments to the barley PFT in ecosys with respect to the wheat PFT were made based on relative performance between ecosys simulated barley and wheat yields, biomass growth, and nitrogen uptake, which were rigorously tested by Grant et al (2020) against Alberta field data. Typical soil, crop, and nutrient management practices and recommended N fertilizer rates across different regions of Alberta were used as model inputs (Figure 2; Table 1).…”