Quantifying the impact of present-day and future tropospheric ozone on crop productivity at global and regional scale using JULES-crop
Leung, Pui Kin Felix
Thesis or dissertation
University of Exeter
Reason for embargo
The chapters of the thesis will be published into papers.
Tropospheric ozone (O3) is the third most important anthropogenic greenhouse gas. It is harmful to human health and detrimental to plant productivity, causing significant crop production losses. Currently, O3 concentrations are projected to increase globally, which could have a significant impact on food security. The Joint UK Land Environment Simulator modified to include crops (JULES-crop) is used here to quantify the impacts of present-day and future tropospheric O3 on crop production at the regional scale until 2100. These regions include the main crop producing countries. We evaluate JULES-crop against the Soybean Free-Air-Concentration-Enrichment (SoyFACE) experiment in Illinois, USA. Experimental data from SoyFACE and various literature sources is used to calibrate the parameters for soybean and ozone damage parameters in soybean in JULES-crop. The calibrated model is then applied to a transient factorial set of JULES-crop simulations over 1960-2005. Modelled yield is then compared with FAO observed yield and model performance is evaluated using Pearson correlation statistics. Yield loss and economic loss of crops due to ozone damage are evaluated. Irrigation and yield gap factors (the difference between observed and potential yields at the same location) are applied to the simulations to investigate the effect of inclusion on model performance. IPCC future climate scenarios RCP 2.6 and RCP 8.5 are employed to simulate the future impact of ozone damage on crops. Simulated yields are compared with those for the year 2050 projections from FAO. Simulated yield changes are attributed to individual environmental drivers, CO2, O3 and climate change, across regions and for different crops. A mixed scenario of RCP 2.6 and RCP 8.5 ozone and climatology, respectively, are used to explore the implication of clean air policy and climate change mitigation policy. Results show that regions with high O3 concentration such as China and India suffer the most from O3 damage; soybean is more sensitive to O3 than other crops, maize is not sensitive to O3, or to CO2 because it is a C4 crop with leaf physiology that limits the benefit from the atmospheric CO2 increase. JULES-crop predicts CO2 fertilisation would increase the productivity of vegetation. This effect, however, is offset by the negative impacts of tropospheric O3. Using data from FAO and JULES-crop, it is estimated that O3 damage has cost around 55.4 Billion USD (at today’s prices) per year for soybean. The Pearson correlation of modelled yield and observation also suggest that JULES is more sensitive to precipitation than temperature. Application of yield gap factor also improves the model performance regarding inter-annual variation. Irrigation improves the simulation of rice only, and it increases the O3 damage because drought can reduce the O3 flux entering through the plant stomata. Future climate scenario simulations show that RCP 8.5 results in a high yield for all crops mainly due to the CO2 fertilisation effect. When this effect is not included O3 damage is large in Asian countries. Mixed climate scenario simulations suggest that RCP 8.5 CO2 concentration and RCP 2.6 O3 concentration result in the highest yield. JULES-crop is still in the early stage of development and subsequently, has missing processes and can be further improved. Further work to use data from more crop FACE-O3 experiments and more crop functional types in JULES are necessary. The model will thus contribute to the understanding of the impacts of climate change on food production. In future work, JULES will later be coupled with the Unified Model to quantify the impact of tropospheric O3 on crops productivity including feedbacks between the land-surface, atmospheric chemistry and climate change.
PhD in Geography
- Doctoral Theses