A genotype, environment and management (GxExM) analysis of adaptation in winter wheat to climate change in Denmark

https://doi.org/10.1016/j.agrformet.2013.11.009Get rights and content

Highlights

  • We model winter wheat yields under climate change with changes in genetics and management for adaptation.

  • We examine trends in yield and suitability and uncertainties of adaptations.

  • Climate change scenarios lead to yield increase in the medium term and reductions by the end of 21st century.

  • Crop development exhibits the highest sensitivity to climate change and the largest adaptation potential through genetic change.

  • Interactions between adaptation options make a combined adaptation assessment and multi-target breeding program suitable.

Abstract

Wheat yields in Europe have shown stagnating trends during the last two decades, partly attributed to climate change. Such developments challenge the needs for increased production, in particular at higher latitudes, to meet increasing global demands and expected productivity reductions at lower latitudes. Climate change projections from three General Circulation Models or GCMs (UKMO-HadGEM1, INM-GM3.0 and CSIRO-Mk3.1) for the A1FI SRES emission scenario for 2000 to 2100 were downscaled at a northern latitude location (Foulum, Denmark) using LARS-WG5.3. The scenarios accounted for changes in temperature, precipitation and atmospheric CO2 concentration. In addition, three temperature-variability scenarios were included assuming different levels of decreased temperature variability in winter and increased in summer. Crop yield was simulated for the different climate change scenarios by a calibrated version of AFRCWHEAT2 to model several combinations of genotypes (varying in crop growth, development and tolerance to water and nitrogen scarcity) and management (sowing dates and nitrogen fertilization rate). The simulations showed a slight improvement of grain yields (0.3–1.2 Mg ha−1) in the medium-term (2030–2050), but not enough to cope with expected increases in demand for food and feed. Optimum management added up to 1.8 Mg ha−1. Genetic modifications regarding winter wheat crop development exhibit the greatest sensitivity to climate and larger potential for improvement (+3.8 Mg ha−1). The results consistently points towards need for cultivars with a longer reproductive phases (2.9–7.5% per 1 °C) and lower photoperiod sensitivities. Due to the positive synergies between several genotypic characteristics, multiple-target breeding programmes would be necessary, possibly assisted by model-based assessments of optimal phenotypic characteristics.

Introduction

Europe is one of the major global suppliers of food and fibre. For wheat, one of the most produced cereals worldwide, European production constitutes 32% of the total (FAOSTAT, 2011). Recent analysis of European cereal productivity show a stagnation of wheat yields (Supit et al., 2010, Olesen et al., 2011, Lin and Huybers, 2012). Such leveling of crop yields, more intensively experienced in southern and central European regions, has been linked to increases of temperatures or more frequent droughts due to climate change (Supit et al., 2010, Olesen et al., 2011, Lin and Huybers, 2012). Yield stagnation has also been observed in higher latitude European countries, where modest changes in climate are expected to favour crop production (Olesen and Bindi, 2002). In these cases, slow or no yield progress has been mostly attributed to reduced management intensities influenced by modifications of agricultural policies that reward other targets than agricultural production and to environmental legislation that limits pesticide and fertilizer applications (Lin and Huybers, 2012). A deeper assessment of Danish agriculture (Petersen et al., 2010) showed that the main causes of stagnating wheat yields were farm management practices (application of animal slurry or subsoil compaction), governmental restrictions on nitrogen (N) fertilization and changes in pest and diseases, partly caused by changes in crop sequences. Conversely, breeding contributed to enhancing wheat productivity.

Current yield stagnation in Scandinavia may threaten the hopes pinned on high latitude countries to compensate for production losses in lower latitude lands from climate change (Parry et al., 1999, Devereux and Edwards, 2004, Lee, 2009). Current global carbon dioxide emissions are following the highest-case scenario from the IPCC 4th Assessment Report (Peters et al., 2012). At the same time, it is important for Europe to remain competitive and active in food production given expected population growth and changes in consumption patterns. This makes it crucial to analyze strategies to overcome current yield limitations to face future food production and food security challenges.

Process-based crop models are tools frequently used to examine impacts, make predictions and evaluate crop management practices. They provide mathematical descriptions of the interactions between environments (E), genotypes (G) and management (M) practices to understand climate impacts and thereafter offer possible adaptations to offset climatically driven drawbacks to crop yields. In agronomic terms, adaptations are arrived at either by breeding new crops or crop varieties or by changing crop management or both (i.e. GxExM).

As models are simplifications of reality, uncertainties (defined as the lack of confidence about specific outcomes of an event) may arise from different sources along the uncertainty cascade of adaptation assessments; from emission scenarios via climate models and the downscaling of weather to be applied as input crop and impact models (Challinor et al., 2012) and from the modeling and handling of adaptation issues (Refsgaard et al., 2012). Uncertainties may increase with larger extrapolations, either in environmental conditions or adaptation measures. Nonetheless, they may give an idea of the most important and sensitive adaptation measures and pointing at directions to follow.

Studies in adaptation analysis have mostly investigated interactions between E and M or E and G (Olesen and Bindi, 2002, Torriani et al., 2007, Olesen et al., 2007, Thaler et al., 2012, Lehmann et al., 2013). Some exceptions (e.g. Ghaffari et al., 2002, Torriani et al., 2007, Luo et al., 2009) changed thermal time requirements, vernalisation and photoperiod responsiveness of the crop in conjunction with changes in sowing dates and nitrogen fertilization. The latter studies examined only a limited spectrum of the entire range of potential agronomic adaptation possibilities.

The objective of this study is to illustrate and analyze transient trends of a wide range of adaption alternatives to climate change at a high latitude location for winter wheat (Triticumaestivum L.), whilst capturing as much uncertainty as possible. Specific aims were (1) to evaluate the effect of changes in climate on winter wheat yield in Denmark and (2) to perform a sensitivity analysis of combined adaptations in management of a winter wheat genotype in short time step projections. We also (3) tracked the changes in GxM interactions associated with a given grain yield target and (4) analyzed the influence of propagation of climate change uncertainties in our adaptation assessment. To achieve those targets, AFRCWHEAT2 was run using a combination of GxExM scenarios. The environmental scenarios were constituted by climate projections from three GCMs under the SRES A1FI emission scenario and three temperature-variability scenarios on a ten year time step. Adaptation scenarios (GxM) were established by systematic changes in management and crop phenotypes involving crop growth, crop development and tolerance to water and nitrogen scarcity.

Section snippets

Methodology

The AFRCWHEAT2 (Porter, 1993) crop simulation model for winter wheat was calibrated and evaluated using two independent data sets. The model was run for nine climate change scenarios with modified temperature, precipitation, temperature variability and increased CO2. Climate change projections were made from three General Circulation Models (GCMs) applying the SRES A1FI emission scenario. MAGICC/SCENGEN (Fordham et al., 2012) was used to emulate global climate conditions over the 21st century

Model calibration and evaluation

Phenology showed a high capability to reduce the scatter of the error of the simulated yield, so among the vectors that provided with better approximations to the observed dates of crop stages (focusing on flowering dates around 1–15th of June and maturity at the first half of August) a low σe was prioritized over the RMSE for the search of the best phenological parameter vector. Several phenology parameter sets accomplished a low σe (<1.3 Mg ha−1), while keeping a low value of the RMSE (RMSE <1.4

Pathway for food security

Food security is expected to rely on increases of crop production rather than on enlargement of crop production areas (Godfray et al., 2010, Ray et al., 2013). The estimated average increase in crop production necessary to feed 9 billion people in 2050 is around 2–4% per year (Ray et al., 2013). This demand has to be satisfied through intensification of crop production, much of which would happen in cool temperate regions such as north-western Europe (Elsgaard et al., 2012). Due to the positive

Conclusions

Moderate changes in climate were simulated to give increases of winter wheat yields at a high latitude location in Europe (Foulum, Denmark). Crop responses to warmer temperatures under current GxM conditions raised grain yields by 0.3–1.2 Mg ha−1in the medium term (2030–2050). This increase is below the theoretical threshold required to feed a population of 9 million in 2050 (that needs a 50% increase in yield). Changes in crop management provided an average yield increase of 1.8 Mg ha−1, which is

Acknowledgements

The study was part of the Centre for Regional Change in the Earth System (CRES—www.cres-centre.dk) under contract no: DSF-EnMi 09-066868 funded by the Danish Strategic Research Council. We thank researchers collaborating in the ADAPTAWHEAT project (contract no: FP7-KBBE-2011-5/289842) for their comments and suggestions.

References (56)

  • J.E. Olesen et al.

    Impacts and adaptation of European crop production systems to climate change

    Eur. J. Agron.

    (2011)
  • M. Özdoğan

    Modeling the impacts of climate change on wheat yields in Northwestern Turkey

    Agric. Ecosyst. Environ.

    (2011)
  • M. Parry et al.

    Climate change and world food security: a new assessment

    Global Environ. Change

    (1999)
  • J. Porter

    AFRCWHEAT2: a model of the growth and development of wheat incorporating responses to water and nitrogen

    Eur. J. Agron.

    (1993)
  • I. Supit et al.

    Recent changes in the climatic yield potential of various crops in Europe

    Agric. Syst.

    (2010)
  • F. Tardieu et al.

    Designing crops for new challenges

    Eur. J. Agron.

    (2012)
  • V. Alekseev et al.

    Modelling of the present-day climate by atmosferic model of INM RAS DNM GMC

    Prepr. INM

    (1998)
  • K. Beven

    Environmental modelling: an uncertain future?

    (2008)
  • C.D. Børgesen et al.

    A probabilistic assessment of climate change impacts on yield and nitrogen leaching from winter wheat in Denmark

    Nat. Hazards Earth Syst. Sci.

    (2011)
  • A.J. Challinor et al.

    Use of agro-climate ensembles for quantifying uncertainty and informing adaptation

    Agric. For. Meteorol.

    (2012)
  • S. Devereux et al.

    Climate change and food security

    IDS Bull.

    (2004)
  • L. Elsgaard et al.

    Shifts in comparative advantages for maize, oat and wheat cropping under climate change in Europe

    Food Addit. Contam. Part A

    (2012)
  • D.A. Fordham et al.

    Strengthening forecasts of climate change impacts with multi-model ensemble averaged projections using MAGICC/SCENGEN 5.3

    Ecograph

    (2012)
  • M. Foulkes et al.

    Genetic progress in yield potential in wheat: recent advances and future prospects

    Can. J. Agric Sci.

    (2007)
  • K. Garrett et al.

    Complexity in climate-change impacts: an analytical framework for effects mediated by plant disease

    Plant Pathol.

    (2011)
  • A. Ghaffari et al.

    Climate change and winter wheat management: a modelling scenario for south-eastern England

    Clim. Change

    (2002)
  • H.C.J. Godfray et al.

    Food security: the challenge of feeding 9 billion people

    Science

    (2010)
  • H. Gordon et al.

    The CSIRO Mk3 Climate System Model

    (2002)
  • Cited by (55)

    • Yield sustainability of winter wheat under three limited-irrigation schemes based on a 28-year field experiment

      2022, Crop Journal
      Citation Excerpt :

      In response to this pressure, there is increasing focus on “sustainable intensification” [8,9]. This strategy for meeting the demand for food security is “expected to rely on increases of crop production rather than on enlargement of crop production areas” [10]. In the NCP, with the aim of increasing crop yields and address water scarcity, the three limited-irrigation schemes (W1, W2, and W3) have been proposed and applied for many years and at many sites.

    • Expected effects of climate change on the production and water use of crop rotation management reproduced by crop model ensemble for Czech Republic sites

      2022, European Journal of Agronomy
      Citation Excerpt :

      The possible effect of cultivar selection or development in the future was not assessed in this study, as constant cultivars were used. On the other hand, there is also some potential, as Montesino-San Martín et al. (2014) estimated using a crop growth model for certain genotype, environment and management conditions reported for Foulum, Denmark. Under climate change for forthcoming decades, genetic modifications of winter wheat exhibit large potential for improvement (+ 3.8 t/ha) compared with that of management (up to +1,8 t/ha).

    • Genomic-assisted breeding for abiotic stress tolerance in horticultural crops

      2021, Stress Tolerance in Horticultural Crops: Challenges and Mitigation Strategies
    View all citing articles on Scopus
    View full text