Simulating agriculture in the Community Land Model version 5

Authors: LOMBARDOZZI, DANIA - National Center For Atmospheric Research (NCAR); LU, YAQIONG - Chinese Academy Of Sciences; LAWRENCE, PETER - National Center For Atmospheric Research (NCAR); LAWRENCE, DAVID - National Center For Atmospheric Research (NCAR); SWENSON, SEAN - National Center For Atmospheric Research (NCAR); OLESON, KEITH - National Center For Atmospheric Research (NCAR); WIEDER, WILLIAM - National Center For Atmospheric Research (NCAR); Ainsworth, Elizabeth - Lisa

Submitted to: Journal of Geophysical Research-Biogeosciences
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/13/2020
Publication Date: 6/22/2020
Citation: Lombardozzi, D.L., Lu, Y., Lawrence, P., Lawrence, D.M., Swenson, S., Oleson, K.W., Wieder, W.R., Ainsworth, E.A. 2020. Simulating agriculture in the Community Land Model version 5. Journal of Geophysical Research-Biogeosciences. https://doi.org/10.1029/2019JG005529.
DOI: https://doi.org/10.1029/2019JG005529

Interpretive Summary: While agricultural management practices are critical for increasing global food production, there is limited understanding of how they impact fluxes of carbon, water, and energy from the land surface to the atmosphere. Global land models are useful for understanding these possible climate impacts, yet few global land models include representations of crops or crop management due to the complexity of the interactions between human decisions and biological processes on global scales. Our analysis illustrates that representing specific crop types, as well as irrigation and fertilization, in the Community Land Model (CLM) increases the amount of carbon that plants draw out the atmosphere and also changes patterns of evaporation. Additionally, crop yield estimates from CLM compare well to observed crop yields until ~1990, when modeled crop yields level off. The difference between observed and modeled yields starting ~1990 is in part because CLM does not model management practices associated with intensification. Overall, our results illustrate the impact that crop management may have on climate and highlights that global models should represent specific crop types and crop management to not accurately capture carbon, water, and energy fluxes from the land surface.

Technical Abstract: Agricultural management has greatly increased global food production and altered Earth’s climate by changing physical and biogeochemical properties of terrestrial ecosystems. Few Earth system models represent agricultural management practices due to the complexity of the interactions between human decisions and biological processes on global scales. We describe the new capabilities of representing crop management in the Community Land Model (CLM) version 5, which includes time-varying spatial distributions of major crop types and their management through fertilization and irrigation, and temperature-based phenological triggers. Including active crop management increases peak growing season GPP (>2 g C m-2 day-1 in crop regions), increasing the amplitude of northern hemisphere net ecosystem exchange, and changes seasonal and annual patterns of latent and sensible heat fluxes. The CLM5 crop model simulates the global observed historical trend of crop yields for the crops represented with relative fidelity from 1850 to 1990. Cropland expansion is important for increasing crop yields, especially during the first century of the simulations, while fertilization and irrigation are important for increasing yields from 1950 onward. From 1990 to present-day, observed crop yields continue to increase while CLM5 yields level off, likely because intensification practices are not represented. Specifically, CLM does not currently include representations of tillage, increasing planting density, or other management practices that may also affect crop-climate and crop-carbon cycle interactions and alter trends in yields. These results highlight the importance of including crop management in Earth system models, particularly as global datasets for parameterization and evaluation become more readily available.