IAP-26-027

Environmental Drivers of Elephants Movement, Behaviour, and Spatial Adaptation in a Changing Savanna Ecosystem

Movement is a primary mechanism through which animals respond to spatial and temporal variation in resources, environmental conditions and ecological disturbance. Understanding how environmental variability translates into movement decisions and behavioural change is therefore a major challenge in ecology, particularly as environmental variability and extreme climatic events increasingly alter the distribution and predictability of resources [1]. However, movement responses emerge not only from current environmental conditions, but also from animals’ previous experience of landscapes and the social context in which decisions are made.

Previous studies have demonstrated relationships between elephant movement and individual environmental drivers and have independently investigated behavioural states, spatial memory or social organisation [2–8]. However, these mechanisms have rarely been integrated within a common analytical framework that include near real-time environmental data. We still have limited understanding of 1) How environmental conditions influence both individual movement steps and broader behavioural states, 2) How previous landscape experiences affects subsequent movement decisions, and 3) Whether social characteristics modify animal responses to movement. Long-term integration of movement trajectories with environmental datasets, ecological and field-derived information provides a unique opportunity to address these challenges.

This project will investigate these interacting mechanisms using African savanna elephants (Loxodonta africana) as a model system. Elephants are highly mobile, long-lived and socially complex herbivores whose movements are strongly influenced by water availability, rainfall, vegetation productivity and drought [2–4]. Their extensive spatial knowledge and complex social organisation also make them particularly suitable for investigating how environmental information, spatial memory and collective decision-making combine to shape movement through time [5–7]. The project will make use of long-term GPS telemetry and environmental monitoring from Selati Game Reserve, South Africa, where strong seasonal and interannual environmental variability provide a natural system for studying behavioural responses to environmental change.
The project will develop an integrated environmental–movement framework combining high-resolution elephant trajectories with environmental data, movement analytics, behavioural modelling and targeted field observations. The main emphasis will be placed on spatial-memory-based movement decisions by linking historical landscape conditions and repeated resource use with subsequent movement behaviour. Field observations will provide ecological context for these patterns and will be integrated with the movement and environmental datasets through spatial data-science approaches. The research questions are:

RQ1: How do environmental conditions and their temporal variability influence elephant movement decisions and behavioural states across seasonal and interannual timescales?
RQ2: To what extent does previous landscape experience predict subsequent elephant movement decisions, route use and resource revisitation after accounting for current environmental conditions?
RQ3: How do herd composition and matriarchal characteristics influence elephant movement decisions and behavioural responses to environmental variability?

The PhD candidate will face key challenges:
1. Integrating heterogeneous remotely sensed, ecological and field datasets operating across different spatial and temporal scales into a unified environmental–movement framework.
2. Developing and applying advanced movement-analytical approaches to distinguish environmental effects on movement trajectories, resource selection and behavioural states.
3. Testing whether previous landscape experience explains subsequent movement decisions beyond contemporary environmental conditions.
4. Designing and conducting targeted fieldwork in South Africa to contextualise and validate movement-derived ecological patterns and integrate field observations within spatial data-science analyses, and support both HMM-state interpretation and ecological characterisation of movement patterns.
5. Integrating environmental, behavioural, spatial-memory and social processes to develop a mechanistic and predictive understanding of elephant responses to environmental change.

The project will be organised through the following stages:
Environmental–Movement Data Framework: Long-term elephant GPS trajectories will undergo quality control and temporal standardisation before deriving movement metrics and integrating each location with spatially and temporally matched environmental information. Environmental datasets from remote sensing sources and reserve-monitoring datasets will characterise dynamic conditions including rainfall, drought, vegetation productivity and water availability, in combination with landscape features such as terrain and reserve road network.
Environmental Drivers of Movement: The integrated trajectories will be used to quantify how environmental variability influences elephant movement across seasonal and interannual timescales. Step-selection approaches will assess how environmental and landscape characteristics influence movement and resource-selection decisions, while multi-individual analyses will identify variation in responses to changing resource conditions and real and extremes environmental conditions. This stage will determine which environmental factors most strongly influence elephant movement and how these relationships vary through time.
Behavioural Responses to Environmental Change: MHH will infer latent behavioural states from movement characteristics, allowing changes in behaviour to be related to environmental conditions. Behavioural-state occupancy and transitions will be modelled across individuals and environmental regimes to determine how elephants modify their behaviour under changing resource availability. Integrating these results with movement-selection analyses will distinguish changes in movement decisions from broader shifts in behavioural state. Where temporally matched field observations are available, directly observed elephant activities (e.g., travelling, foraging and resting) will be compared with HMM-inferred states to evaluate their behavioural interpretation and identify potential misclassification.

Spatial Memory, Field Validation and Social Structure: Revisitation, recursion, residence-time and site-fidelity analyses will characterise repeated space use. To investigate movement patterns consistent with spatial memory, subsequent movement decisions will be modelled against previous landscape experience, including prior visits and historical resource conditions, while accounting for contemporary environmental conditions and resource availability. Fieldwork at Selati Game Reserve will provide ecological characterisation of repeatedly used resources and structured behavioural observations of GPS-collared elephants to support interpretation of movement-derived states. The stage aims to evaluate how social context modifies environmental responses, behavioural-state dynamics and repeated resource use.

Year 1

Literature review and methodological training in movement ecology (tailored to elephants’ behaviour) and spatial data science. Pre-processing, of long-term elephant GPS trajectories with remote sensing environmental datasets, and reserve-monitoring datasets. Derivation of movement metrics and application of movement-selection approaches to identify the principal environmental drivers of elephant movement across individuals, seasons and years. Establishing behavioural-observation protocol and initial HMM comparisons. Initial architecture of the data fusion framework. Initial results will be presented at a national conference (poster or short paper), draft or submission of the first PhD paper.

Year 2

Building on framework identified in Year 1, the project will investigate spatial-memory-based movement decisions using revisitation, recursion, residence-time and site-fidelity analyses. Current movement decisions will be linked to previous environmental conditions and resource distributions to determine whether historical landscape experience predicts subsequent space use and behavioural responses. First fieldwork at Selati Reserve focus on results dissemination, used resources, movement areas and routes identified from the movement analyses, alongside structured behavioural observations of GPS-collared elephants to evaluate HMM-state interpretations and collection of herd composition and matriarchal information. Second PhD paperwork and first International Conference, likely to Movement Ecology where preliminary results can be discussed.

Year 3

Spatial-memory analyses will be extended across additional years, individuals and contrasting environmental conditions to evaluate the persistence and flexibility of memory-informed movement strategies through time. A second fieldwork on the reserve will revisit key resources, movement areas and routes identified during Year 2, while continuing structured behavioural observations to evaluate HMM-state interpretations and collecting additional herd composition and matriarchal information. Movement, behavioural-state, environmental, field and social information will then be integrated to assess how previous landscape experience and social characteristics influence movement responses beyond contemporary environmental conditions. Work on third PhD Paper and international conference to dissemination and feedback.

Year 3.5

Dissertation write-up and finalising the research findings. Dissemination of research through academic publications (excepted at least 3), two national and one international conference presentations across the doctoral period, and knowledge sharing with relevant stakeholders.

This project will build on the candidate’s existing foundations in movement ecology, and elephant telemetry analysis, while developing advanced expertise in spatial and movement data science. Core methodological training will focus on step-selection approaches, multi-individual and longitudinal movement modelling, spatial-memory analysis, and data fusion methods and field datasets. This project will leverage expertise from the University of Glasgow and South Wales in terms of advanced statistical workflows for investigating behavioural and social drivers of animal movement. Field work experience will further strengthen skills in behavioural and ecological data processing, ground-truthing and the integration of field observations with movement-derived information.
This framework will leverage expertise across the supervisory team, and IAPETUS Doctoral Training Partnership, providing access to specialist training in quantitative ecology, spatial analysis, environmental data science and interdisciplinary research. Participation in relevant conferences will further develop expertise in movement analysis, spatial-memory approaches and reproducible ecological modelling, while collaboration with Selati Game Reserve will provide applied training in field ecology.
Further training in key transferable skills is available through The International Education and Lifelong Learning Institute (IELLI). at the University of St Andrews.