Home arrow Scholarships and Job offers arrow A funded PhD Studentship at Northumbria University
A funded PhD Studentship at Northumbria University
martes, 17 noviembre 2020

A PhD Studentship funded by the ONE Planet Doctoral Training Partnership

Project: Understanding the spatial-temporal risk of forest fire at multiscale: a statistical modelling approach
Reference: OP2195

Closing date: Monday, January 18, 2021

Project detail:

Forest fires, when spreading uncontrollably, can be detrimental, causing significant damage to the environment, disrupting and even endangering human lives. Due to climate change, we have already been experiencing an unprecedented number of devastating wildfire events globally. In addition to the climatic factors, increasing human activities (e.g. farming and creating residential settlements) in close proximity to forests poses further risks to the onset of wildfires. Embedded within the NERC key research area of Climate & Climate Change, this project tackles the issue of wildfire through the Anthropocene and Environmental Informatics angles.

There are two parts to this project. The first part aims to reveal the space-time variability of wildfire risk at both global and subnational scale. At the global level, analyses will be carried out across countries to understand the country-level risk and how that risk evolves over time. At the subnational scale, interests lie in identifying local wildfire hotspots and monitoring the local temporal risk patterns. This project will develop a novel Bayesian multilevel, multivariate space-time frame work that(a) incorporates the multiple nested geographical levels (e.g., areas within countries which are nested within continents) at which wildfire data are georeferenced and (b) quantifies wildfire risk through multiple dimensions, e.g. number of fire events, burned area and duration. Models of this type are discussed in Haining and Li (2020;ISBN-10:1482237423).

Building upon the first part, the second part investigates how the global, national and subnational space-time patterns are influenced by climatic, topographic and human-activity factors. Of particular interest is to assess how long-term (annual or decadal)changes in these factors affect the change in wildfire risk. This allows us to anticipate future risk patterns under different climate change scenarios and/or different scenarios of how humans interact with forests. At the subnational, shorter-term scale, we will focus on questions such as what factors are affecting risks in a specific locality and how many wildfire events will occur in the next month. These questions are directly relevant to inform policies for local wildfire management. With the increasing availability of data on individual wildfire events (e.g.,the Global Wildfire Information System), space-time point pattern models will also be employed to understand micro-scale wildfire properties, e.g., how wildfire spreads under different causes (e.g.,naturally occurring vs human caused) and different meteorological and topographic conditions.

Requirement: A Bachelor (preferably a Master’s) degree in a statistical subject. Knowledge of modelling and managing data in R/Python and experience in modelling spatial and spatial-temporal data are desirable.

Funding Notes: This studentship is funded by the Natural and Environmental Research Council (NERC) through the ONE Planet Doctoral Training Partnership. The studentship includes 3.5 years of fees (Home), an annual living allowance (£15,285) and a Research Training Support Grant (for travel, consumables, as required). Home and International students (inc. EU) are welcome to apply. A limited number of International awards are available, and applicants may be required to provide additional fees not covered by the UKRI Home fee contribution.

How to Apply: How to Apply | ONE Planet Doctoral Training Partnership. Submit your application through the "Apply to Northumbria" link at the bottom of the page.

For any inquiry, please contact Dr. Guangquan Li at This e-mail address is being protected from spam bots, you need JavaScript enabled to view it .