Assessing alfalfa production under historical and future climate in eastern Canada: DNDC model development and application

https://doi.org/10.1016/j.envsoft.2019.104540Get rights and content

Highlights

  • Alfalfa biomass and yield were used to develop improved alfalfa growth functions in DNDC.

  • Integrated winterkill effects including hardening and dehardening with cultivar specificity.

  • Improved DNDC model showed lower bias in biomass than pre-development version.

  • Yields increased notably under future climate due to early planting and more cuts.

  • Winterkill incidence remained a problem under future scenarios due to less snow cover.

Abstract

The DeNitrification-DeComposition (DNDC) model was developed to better simulate alfalfa (Medicago sativa L.) production with winterkill effects in eastern Canada. The pre-development DNDC produced “fair” simulations of alfalfa yield and biomass (index of agreement (d) > 0.7, Nash-Sutcliffe efficiency (NSE) > 0) but normalized root mean square error (nRMSE) was > 30% whereas the improved model indicated “good” to “excellent” performance (nRMSE = 3.9–29%; d = 0.93–0.99; NSE = 0.76–0.99). Under future RCP4.5 and 8.5 scenarios, average annual yields increased by 60.3% and 81.8% respectively due to earlier planting/spring regrowth with additional cuttings, increased C assimilation and reduced water stress under higher CO2 concentrations. For locations at Ottawa and Quebec City there could be an increased incidence of winterkill under future climate due to reduced snow cover which leads to colder soil/crown temperatures as well as reduced fall hardening. This study indicated that the use of winter hardy cultivars could mitigate winterkill effects and increase production.

Introduction

Alfalfa (Medicago sativa L.) is the most widely grown forage legume worldwide and plays a key role in the development of sustainable cropping and livestock production systems (Crews and Peoples, 2004; Castonguay et al., 2006; Thivierge et al., 2016). Alfalfa contributes to sustainable agricultural systems by improving soil quality, increasing water infiltration and soil water storage, reducing soil erosion, and reducing nutrient losses (e.g., N2O fluxes and NO3 leaching) (Sarrantonio and Gallandt, 2003; Confalonieri and Bechini, 2004; Blackshaw et al., 2010). There is considerable interest in maximizing economic return and minimizing environmental cost for cropping systems used in dairy livestock production where the alfalfa crop is often an intergral component (Confalonieri and Bechihi, 2004; Brown, 2009). In eastern Canada, perennial forage crops account for about 40% of total agricultural land (2.1 million ha) with an annual estimated farm value of 1.3 billion Can$ (Statistics Canada, 2001; Bélanger et al., 2006). It has been shown that the yield and N content of wheat following alfalfa was increased compared to wheat following a canola control (Bullied et al., 2002). Compared to monocultural corn, alfalfa in rotation with corn increased soil organic carbon, increased corn yields, reduced N fertilizer rate and decreased N2O emissions from agricultural fields (MacKenzie et al., 1998; Gregorich et al., 2001). In addition, Jarecki et al. (2018) noted that for corn rotations that included alfalfa, an increase in soil carbon and corn yields was observed as compared to monoculture corn and that these increases continued under future climate scenarios whereas in the monoculture system no further change was realized.

A major challenge for perennial crops in northern regions is the harsh overwintering conditions that contribute to frequent stand losses and yield reduction in Canada (Suzuki, 1972; Ouellet, 1976; Bélanger et al., 2006), several regions of the United States (Barnhart et al., 2004; Castonguay et al., 2006), and Northern Europe (Eagles et al., 1997). The winterkill of perennial crops is largely affected by climatic conditions including subfreezing temperatures, lack of snow cover, and ice encasement (Ouellet, 1976; Schwab et al., 1996; Bélanger et al., 2006). Cold tolerance of cultivars is the capacity of plants to tolerate subfreezing temperatures which can significantly influence the winter survival of perennial crops (Bélanger et al., 2002; Castonguay et al., 2011). Previous studies assessed the effects of environmental factors on the survival of winter-sensitive forage species including alfalfa under both natural and controlled conditions (Paquin and Mehuys, 1980; McKenzie and McLean, 1984; Schwab et al., 1996; Bélanger et al., 2014). The reported correlations between climate variables and winter damages were used to explore optimal climatic conditions (e.g., temperature, precipitation) for winter survival (Ouellet, 1976; Bélanger et al., 2002; Castonguay et al., 2011), and to develop appropriate algorithms for modelling winter hardiness and injury (Kanneganti et al., 1998a, b).

Changes in temperature, precipitation, and atmospheric CO2 concentration influence crop growth by impacting the rate of photosynthesis, respiration, and crop water use efficiency (WUE) along with influencing soil biological and chemical transformations of C and N (Guo et al., 2010; Wang et al., 2014; Long et al., 2015). Alfalfa, a C3 legume species, benefits from elevated atmospheric CO2 due to increased photosynthesis (Morgan et al., 2004) and improved water use efficiency (Leakey et al., 2009; Barton et al., 2011). Many forage species, including alfalfa, are sensitive to drought, and would be affected by future changes in precipitation rate and pattern (Aranjuelo et al., 2007; Thivierge et al., 2016). The positive response to increased CO2 concentration might be offset by an increased temperature, increased evapotranspiration or reduced precipitation (Hatfield et al., 2011; Lee et al., 2013; Piva et al., 2013). Climate change may therefore have greatly varying effects on alfalfa production depending on the selected cultivars, management, and soil and climatic conditions (Izaurralde et al., 2011; IPCC, 2012; Thivierge et al., 2016).

Process-based models have been widely employed across diverse agroecosystems to estimate crop production, hydrologic processes, and nutrient cycling (Jones et al., 2003; Williams et al., 2008; Li et al., 2012; Holzworth et al., 2014). The DeNitrification-DeComposition (DNDC) model is an example of a well-established tool used to simulate crop growth and development, GHG emissions and water quality in crop, livestock and forest systems (Li et al., 1992a, b; Smith et al., 2013; Congreves et al., 2016). The DNDC model (Li et al., 1992a, b) was initially developed to predict greenhouse gas (GHG) emissions from agricultural soils and was expanded to simulate soil C&N cycling and nitrate leaching (Li et al., 2006) and more recently to simulate nutrient cycling for full farm facility and livestock systems (Li et al., 2012). The simulation of crop growth has advanced over the last 25 years (Zhang and Niu, 2016) and certain versions of DNDC use detailed approaches for simulating physiological processes such as Crop-DNDC (Zhang et al., 2002). Several countries have developed regional versions of DNDC to improve the predictions of their agricultural systems (Brown et al., 2002; Leip et al., 2011; Smith et al., 2013; Gilhespy et al., 2014; Li et al., 2017) including a Canadian version which has been used to simulate crop, soil and plant processes under cool climate conditions (Kröbel et al., 2011; Smith et al., 2013; Dutta et al., 2016; He et al., 2018). Currently, the model algorithms have been calibrated and validated to characterize Canadian growing conditions including sub-models for crop growth (Kröbel et al., 2011), soil organic C dynamics (Smith et al., 2012), N2O emission (Uzoma et al., 2015), evapotranspiration (Dutta et al., 2016), NH3 volatilization (Congreves et al., 2016), and soil temperature (Dutta et al., 2018). Much of the development work has been conducted through comparison with other modeling frameworks, in particular with DayCent, DSSAT and RZWQM2 (Grant et al., 2016; Dutta et al., 2017; Guest et al., 2017; Smith et al., 2019) along with the cooperation with the primary US developers of DNDC.

The current DNDC model, however, is limited in its ability to simulate perennial crop growth, especially regrowth after cutting and in subsequent seasons considering winterkill impacts. This limits the application of the model for exploring scenarios under livestock-based systems that include forages. Therefore, the objectives of this study were to (1) improve alfalfa growth and development in the DNDC model using measured biomass and yield data under various soil and climate conditions; (2) compare performance of the revised and default versions for simulating long-term alfalfa production, soil temperature and moisture and; (3) assess projected climate change impacts on alfalfa production including winterkill in eastern Canada.

Section snippets

Research site and experiment

Measurements of aboveground biomass during the growth cycles at one location in eastern Canada were used independently to develop a new phenological growth curve and then validate this new growth curve. Alfalfa was seeded in 2014 at the Canadian Food Inspection Agency (CFIA) greenbelt farm southwest of Ottawa (45°18′N, 75°45′W) (Fig. 1). Aboveground biomass samples were taken between 2 and 8 times in each of three growth cycles in the seeding year, and four growth cycles in the first (2015) and

Aboveground biomass during growth cycles

The calibration of the default DNDC model resulted in “good” or “fair” agreements between the simulated and measured alfalfa aboveground biomass based on the values of NSE > 0 and d index ≥ 0.8, but the alfalfa aboveground biomass of the spring regrowth was underestimated in 2015 (first post-seeding year) and 2016 (second post-seeding year) (Fig. 2 and A.1, Table 3). The improved DNDC model performed very well in simulating alfalfa aboveground biomass for all years with NSE > 0.9, d > 0.9, and

Conclusions

The DNDC model was modified to improve alfalfa growth simulation using measured biomass and yield data in eastern Canada. The results demonstrated that the improved model with enhanced alfalfa growth algorithms showed statistically better performance compared to the default model in most biophysical variables. The simulated soil temperature had “good” agreements with the measured values with better simulation for improved DNDC. In addition, the DNDC model predicted soil water contents well in

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The authors acknowledge the financial support of Science and Technology Branch of Agriculture and Agri-Food Canada. We also gratefully thank Ray Desjardins for guidance and Devon Worth for the technical support.

References (94)

  • B. Dutta et al.

    Model development in DNDC for the prediction of evapotranspiration and water use in temperate field cropping systems

    Environ. Model. Softw.

    (2016)
  • S.L. Gilhespy et al.

    First 20 years of DNDC (DeNitrification DeComposition): model evolution

    Ecol. Model.

    (2014)
  • R. Guo et al.

    Responses of crop yield and water use efficiency to climate change in the North China Plain

    Agric. Water Manag.

    (2010)
  • W. He et al.

    Estimating the impacts of climate change on crop yields and N2O emissions for conventional and no-tillage in Southwestern Ontario, Canada

    Agric. Syst.

    (2018)
  • D.P. Holzworth

    APSIM–Evolution towards a new generation of agricultural systems simulation

    Environ. Model. Softw.

    (2014)
  • G. Jégo et al.

    Calibration and performance evaluation of the STICS crop model for simulating timothy growth and nutritive value

    Field Crop. Res.

    (2013)
  • J.W. Jones et al.

    The DSSAT cropping system model

    Eur. J. Agron.

    (2003)
  • A. Leip et al.

    Estimation of N2O fluxes at the regional scale: data, models, challenges

    Curr. Opin. Environ. Sustain.

    (2011)
  • C. Li et al.

    Modeling nitrate leaching with a biogeochemical model modified basedon observations in a row-crop field in Iowa

    Ecol. Model.

    (2006)
  • H. Li et al.

    The development of China-DNDC and review of its applications for sustaining Chinese agriculture

    Ecol. Model.

    (2017)
  • S. Liu et al.

    Modelling crop yield, soil water content and soil temperature for a soybean-maize rotation under conventional and conservation tillage systems in Northeast China

    Agric. Water Manag.

    (2013)
  • J.E. Nash et al.

    River flow forecasting through conceptual models part I-a discussion of principles

    J. Hydrol

    (1970)
  • X. Qin et al.

    A long-term sensitivity analysis of the denitrification and decomposition model

    Environ. Model. Softw

    (2013)
  • W.N. Smith et al.

    Crop residue removal effects on soil carbon: measured and inter-model comparisons

    Agric. Ecosyst. Environ.

    (2012)
  • W.N. Smith et al.

    Assessing the effects of climate change on crop production and environmental sustainability at long-term experimental sites in Canada

    Agric. Ecosyst. Environ.

    (2013)
  • W. Smith et al.

    Comparing hydrological frameworks for simulating crop biomass, water and nitrogen dynamics in a tile drained soybean-corn system: cascade vs computational approach

    J. Hydrol

    (2019)
  • K.C. Uzoma et al.

    Assessing the effects of agricultural management on nitrous oxide emissions using flux measurements and the DNDC model

    Agric. Ecosyst. Environ.

    (2015)
  • W. Wang et al.

    Responses of rice yield, irrigation water requirement and water use efficiency to climate change in China: historical simulation and future projections

    Agric. Water Manag.

    (2014)
  • J. Yang et al.

    Statistical methods for evaluating a crop nitrogen simulation model NABLE

    Agric. Syst.

    (2000)
  • Y. Zhang et al.

    The development of the DNDC plant growth sub-model and the application of DNDC in agriculture: a review

    Agric. Ecosyst. Environ.

    (2016)
  • Y. Zhang et al.

    A simulation model linking crop growth and soil biogeochemistry for sustainable agriculture

    Ecol. Model.

    (2002)
  • T.I. Akinseloyin

    Soil Moisture and Temperature Simulation Using the Versatile Soil Moisture Budget Approach

    (2015)
  • S. Asseng et al.

    Uncertainty in simulating wheat yields under climate change

    Nat. Clim. Chang.

    (2013)
  • S. Barnhart et al.

    Estimating Cost of Pasture and Hay Production. Ext. Bull. AG-96

    (2004)
  • C.V.M. Barton et al.

    Effects of elevated atmospheric CO2 on instantaneous transpiration efficiency at leaf and canopy scales in Eucalyptus saligna

    Glob. Chang. Biol.

    (2011)
  • G. Bélanger et al.

    Winter damage to perennial forage crops in eastern Canada: causes, mitigation, and prediction

    Can. J. Plant Sci.

    (2006)
  • G. Bélanger et al.

    Benefits of mixing timothy with alfalfa for forage yield, nutritive value, and weed suppression in northern environments

    Can. J. Plant Sci.

    (2014)
  • G. Bélanger et al.

    Climate change and winter survival of perennial forage crops in eastern Canada

    Agron. J.

    (2002)
  • R.E. Blackshaw et al.

    Suitability of legume cover crop-winter wheat intercrops on the semi-arid Canadian prairies

    Can. J. Plant Sci.

    (2010)
  • A. Bootsma et al.

    Estimates of seeding dates of spring wheat on the Canadian Prairies from climate data

    Can. J. Plant Sci.

    (1988)
  • N.C. Brady et al.

    The Nature and Properties of Soils

    (2002)
  • W.J. Bullied et al.

    Grain yield and N benefits to sequential wheat and barley crops from single-year alfalfa, berseem and red clover, chickling vetch and lentil

    Can. J. Plant Sci.

    (2002)
  • Y. Castonguay et al.

    Cold-induced biochemical and molecular changes in alfalfa populations selectively improved for freezing tolerance

    Crop Sci.

    (2011)
  • Y. Castonguay et al.

    An indoor screening method for improvement of freezing tolerance in alfalfa

    Crop Sci.

    (2009)
  • Q. Deng et al.

    Assessing the impacts of tillage and fertilization management on nitrous oxide emissions in a cornfield using the DNDC model

    J. Geophys. Res. Biogeosci.

    (2016)
  • C.F. Eagles et al.

    Stress physiology and crop improvement

  • Cited by (28)

    • Modelling the impacts of inhibitors and fertilizer placement on maize yield and ammonia, nitrous oxide and nitrate leaching losses in southwestern Ontario, Canada

      2023, Journal of Cleaner Production
      Citation Excerpt :

      The model has been applied successfully in many parts of the world to predict crop growth, soil temperature and moisture regimes, soil C & N dynamics, nutrient loss from leaching and runoff, and emissions of trace gases under different management practices and variable weather conditions (Smith et al., 2013; Deng et al., 2016; Brilli et al., 2017). A modified Canadian version of DNDC has evolved over the past ten years to improve simulation of N2O emissions, NH3 volatilization and N leaching for a wide range of management practices in cool, humid soils (Smith et al., 2013, 2020; Dutta et al., 2016a; b, 2018; Jarecki et al., 2018; He et al., 2019b). The model has been used successfully for delineating climate and management effects on N dynamics in Canadian agroecosystems (Abalos et al., 2016; Banger et al., 2020; Jiang et al., 2021).

    View all citing articles on Scopus
    View full text