IAP-24-031

Use of remote sensing data to improve predictions of ozone and drought impacts on crop yield.

Tropospheric ozone is a harmful pollutant for plant health and can have numerous negative impacts within an ecosystem, including effecting plant growth, biodiversity and carbon sequestration. Globally, ozone pollution compromises food security, with average ozone-induced annual yield losses of 7% for wheat. While the concentration of ozone is decreasing or levelling out in the USA and Europe, increases continue in developing regions. There are also areas of the world where ozone concentration data are not widely available, however ozone is predicted to be causing substantial yield reductions, for example in India.

As ozone pollution presents a considerable future environmental challenge for policy makers, improved understanding of ozone impacts on crops is a priority. An important topic, particularly as the planet faces more extreme weather events, is investigating how ozone interacts with other crop stresses such as drought, and identifying hotspots of high risk where stresses co-occur. Knowledge gaps also remain on ozone concentrations in developing countries and information on crop growth periods and the key drivers of effects on crop yield is needed for calculation of ozone uptake to plants and crop modelling.

This PhD aims to use modelled data and remote sensing to fill these knowledge gaps, working on a global scale and also using more localised case studies, for example in India. Key questions will include ‘can modelled and remote sensing data be used to highlight where ozone and other key stresses co-exist, allowing hotspots to be identified?’, ‘can satellite data improve estimates of timings of key growth stages and crop yield?’, ‘can satellite data be used to estimate ozone impacts in regions that do not have ground-based ozone data, and how do these values compare with modelled data?’ Project results will be used to improve risk assessment of ozone pollution, providing estimates at a large scale, which is particularly useful for areas where measured data are scarce.

The supervisor combination provides complementary expertise for this multidisciplinary project, including extensive experience in investigating ozone impacts on vegetation and using satellite and airborne remote sensing applications for landscape and vegetation characterisation.

The student will process and analyse modelled data (using EMEP and DO3SE models) and gather and process remote sensing data (for example, ozone concentration and vegetation indices). GIS software and R/Python will be used to access and process the datasets. To help fill knowledge gaps on ozone concentration and also potentially to ground truth modelled values, diffusion tubes will be used to measure ozone concentration over 2-4 week periods, situated in rural areas in Uttar Pradesh, Northern India.

The student will also have a placement (~3 weeks) to visit scientists investigating ozone impacts in India (planned for year 2), gathering physiological measurements from ozone exposure experiments and ground truthing crop phenology outputs. There will also be the opportunity to collect experimental data at the UKCEH Bangor solardomes site (including measurements of stomatal conductance). These experiences will give insight into the collection of raw data required for ozone and crop modelling.

The student will benefit from the knowledge and experience of two additional advisors – Dr Katherine Steele, a Senior Lecturer in Sustainable Crop Production (Bangor University), with a background in investigating crop responses to drought stress. Also Dr Richa Rai, Senior Scientist at the CSIR – National Botanical Research Institute (Uttar Pradesh, India), who investigates ozone impacts on the productivity of Indian crops.

Year 1

Interactions between ozone and other stresses (e.g. drought) will be investigated. Ozone stress has been found to co-occur with other key crop stresses. Using modelled and/or remote sensing data, this will be explored further, producing risk maps showing hotspots of crop stress. In addition, the DO3SE model, which estimates ozone uptake by plants, will be used to conduct a sensitivity analysis. Key inputs such as temperature, ozone, water stress, phenology and cultivar, will be varied for selected areas (with data available to mimic realistic conditions), to determine which factors may have the largest influence on final crop yield.

The student will also organise deployment of ozone diffusion tubes at rural sites in India, in collaboration with Indian colleagues researching ozone impacts on crops.

Year 2

The student will investigate how crop growth periods can be determined using remote sensing data, starting with a literature review on the different metrics and methodologies that are available (e.g. NDVI, EVI). Multi-source remote sensing data will be acquired and processed using platforms such as Google Earth Engine to determine crop growing seasons for a case-study region in India. Results will be compared with in-situ ground measurements, along with an investigation of how ozone flux values and impacts on crop yield vary depending on the growth period used.

Phenology data will also be used to model crop yield, along with other environmental variables. The results of this step will be extremely useful for estimating ozone impacts as large scale spatial data on yield are often required. This process has been tested for areas of the UK already – the student will investigate the applicability of this method for other parts of the world. Model inputs will be varied to evaluate the key drivers of effects on crop yield.

The placement to India is planned for year 2 (timing dependent on growing season).

Year 3

Ozone satellite data will be processed (e.g. CAMS dataset (EAC4)), along with ozone concentration data from the EMEP (European Monitoring and Evaluation Programme) model, with the aim of comparing the datasets.

Validation data will also be collated and analysed, for example, data are available for some rural sites in India and a monitoring network across South Africa. Diffusion tube measurements will be used as supporting data.

Year 3.5

Throughout the project, the student will be encouraged to produce manuscripts, and to present their findings from an air quality policy perspective. The last 6 months of the project will be used to complete any final analyses and finalise the PhD thesis.

The programme of research will allow development of key skills required to make the transition from PhD student to independent researcher. These skills will be developed throughout the PhD, from a combination of training courses, one-to-one sessions and mentoring. An initial training needs assessment will be followed by annual reviews of requirements in addition to skills gaps identified during supervision meetings. The student will benefit from postgraduate training schemes available at UKCEH and Bangor University, including training in ethics, health and safety, statistical analysis, quality assurance, presentation and writing skills and research impact.

Skills in specific techniques will be acquired through hands-on training at UKCEH and Newcastle University. This training will be on-going during the project and will include data processing and analysis, GIS mapping and ozone risk assessment, and handling/processing remote sensing data. The student will join the programme of weekly site-wide meetings at UKCEH Bangor and become part of the cross-site ‘Soils and Land Use’ group, allowing discussion of multidisciplinary science ideas and further developing research communication skills. They will have the opportunity to join peer to peer support groups, such as the Remote Sensing User Network (RSUN) at UKCEH.