In her work she developed a method based on sensitivity analysis for assessing impact responses. The method proved to work well in the evaluation of changes in grain yields. According to the results, earlier sowing, and the use of crop cultivars with a short pre-flowering phase followed by a long post-flowering phase would help in securing levels of grain yield. Pirttioja defends her doctoral dissertation at the University of Helsinki on 3 September.
National plans for adapting to climate change require evaluation of risks. As a foundation for risk evaluations, methods are needed that allow the effects of climate change to be examined as reliably as possible.
In her dissertation Nina Pirttioja develops a method based on sensitivity analysis which makes it possible to examine the features of models for evaluating the effects of climate change. The method was applied to models used for evaluating grain yields.
The method involves the formation of impact response surfaces - diagrams using contour lines to depict the sensitivity of modelled yield to changes in temperature and precipitation. The behaviour of different models under changes in climate can be examined by visually examining and comparing contour patterns to detect deviations in results, which can be examined more closely for interpreting and later making use of them.
“Estimates of the impact of climate change are typically bound to selected climate scenarios and revealing uncertainties about those scenarios is a challenge. When using impact response surfaces, the results are not linked with any specific climate scenario. Instead, the method is used to get a broader view of the behaviour of the model in different kinds of climatic conditions”, Nina Pirttioja says.
“Impact response surfaces can then be combined with probability-based climate scenarios, making it possible, for example, to evaluate the likelihood that certain target levels for grain yields could be achieved as climate change advances.”
Examining wheat and barley yields of Finland, Germany, and Spain
The effects of climate change on yields were examined at sites in Finland, Germany, and Spain, where the growing conditions differ from each other. Under examination were 26 crop models for wheat, which were compared with each other, as well as one crop model for barley.
According to the modelling, grain harvests suffer from higher temperatures and lower precipitation levels and benefit from greater precipitation. In Finland, changes in temperature had a greater effect on harvests than precipitation levels did. Spain had opposite results. The results of individual models sometimes deviated considerably from each other.
Model evaluation concluded that the best conditions for increasing crop yield and reducing the probability of crop reduction in the future would come from earlier sowing and selecting cultivars with a short pre-flowering phase, followed by a long post-flowering phase.
Crop models were compared with each other
Key material for research included a study comparing 26 different crop models for wheat. It made it possible to extensively examine how different models react to changes in the climate.
“No model can depict reality perfectly. When the same hypotheses are used for modelling the same thing, different models can easily reach quite different conclusions. In such a case it is important to seek an understanding of the reasons for the differences. It is often very tempting to use the results of just one model as a basis for making decisions. However, it would be important to examine how the results of a model compare with other models that simulate the same thing”, Pirttioja points out.
The dissertation work showed that the impact response surfaces offer a valid alternative to scenario-based methods for evaluating risks caused by climate change and numerous benefits for the analysis of the results and for visual presentation.
“The use of the method is not limited to the examination of crop models. It can be applied to the evaluation of risks connected with e.g. forests, water resources, and health. The evaluations create a more reliable foundation for the planning of measures for adapting to the risks of climate change”, Pirttioja says.
Nina Pirttioja, M.Sc., researcher, Finnish Environment Institute, SYKE
tel. +358 295 251 515, firstname.lastname@example.org
Nina Pirttioja will defend her doctoral dissertation on Friday, 3.9.2021 at 13:00 at the Department of Forest Sciences of the Faculty of Agriculture and Forestry. The event will be in English.