New opportunity

Research Fellow (Tropical Ecosystem Monitoring and Simulations)

Nanyang Technological University · Singapore

About this role

The Yanyan Cheng research group at the Asian School of the Environment (ASE) at Nanyang Technological University (NTU) is seeking a Research Fellow for a one-year, potentially renewable appointment. This role is part of a Singapore NRF-funded project focused on Earth system modeling and understanding interactions among land, water, carbon, energy, atmosphere, ecosystems, and human systems. Responsibilities: Collect, validate, and harmonize field and remote sensing imagery for ecosystem monitoring. Develop machine/deep learning-based workflows to interpret ecosystem disturbance. Synthesize model simulations and multi-source observations to diagnose atmospheric responses to land surface property changes and land–atmosphere feedback. Mentor Research Assistants and students. Requirements: PhD in physical geography, remote sensing, or ecology. Proficiency in scientific computing, high-performance computing (HPC), and scripting (Bash, CDO/NCO). Advanced programming skills in Python, with additional proficiency in R, NCL, and MATLAB. Strong quantitative and analytical skills in working with climate model and observational datasets. Familiarity with remote sensing data processing (satellite, airborne, ground) and sensors (optical, radar, lidar). Experience with tropical climate and environmental studies. Excellent written and oral communication skills in English for stakeholder engagement. Proven track record of publishing in high-quality research journals. Preferred qualifications include experience in ecosystem monitoring, manipulation experiment design, development of diagnosis frameworks (e.g., knowledge-guided machine learning, causal inference, Bayesian neural networks), and evaluation of simulation results such as climate extremes and land-atmospheric interactions.

For full details about this role, click “Apply on company site” to view the employer’s original posting.

Skills for this role

PythonRNCLMATLABMachine LearningDeep LearningKnowledge-guided machine learningCausal inferenceBayesian neural networksRemote sensingScientific computingHigh-performance computingHPCBashCDONCOEarth system modeling

Checking your profile…

Preparation suggestions, not the employer’s actual interview questions. Always check the original posting for current requirements.

Understand the role

What matters most

Development of Advanced Machine Learning Workflows
The role requires creating and implementing machine or deep learning-based workflows to interpret ecosystem disturbances. Candidates must demonstrate the ability to move beyond standard models by applying specialized frameworks like knowledge-guided machine learning, causal inference, or Bayesian neural networks to complex environmental datasets.
Synthesis of Multi-Source Earth System Data
A core duty involves harmonizing field and remote sensing imagery (optical, radar, lidar) with model simulations. Success depends on the ability to diagnose atmospheric responses to land surface changes, requiring deep proficiency in scientific computing and high-performance computing (HPC) environments.
Scientific Communication and Mentorship
The fellow must effectively communicate research findings to stakeholders and contribute to the academic environment by mentoring Research Assistants and students. This requires a proven track record of high-quality publications and the ability to translate complex technical analyses into clear, actionable scientific insights.

Likely challenges to think through

Integrating Heterogeneous Data Streams
The requirement to harmonize diverse data from satellite, airborne, and ground sensors with climate model simulations presents a significant technical hurdle. Prepare by highlighting specific experience in data cleaning, validation, and the use of tools like CDO/NCO to manage large-scale climate datasets.
Balancing Computational Complexity with Scientific Rigor
Developing advanced diagnostic frameworks while ensuring they remain physically grounded in tropical climate dynamics is difficult. Be prepared to discuss how you validate machine learning outputs against physical climate models to ensure the results are scientifically sound rather than just statistically correlated.

Practice your answers

Technical practice

Integrating Multi-Source Remote Sensing Data

What to coverDiscuss your experience in harmonizing optical, radar, and lidar datasets for ecosystem monitoring. Explain how you handle spatial and temporal resolution discrepancies when validating field imagery against satellite or airborne observations.

Machine Learning for Ecosystem Disturbance

What to coverDescribe a workflow you have developed to interpret ecosystem disturbances using deep learning. Address how you ensure model generalizability across different tropical environments and how you validate the results against ground-truth data.

Knowledge-Guided Machine Learning (KGML)

What to coverExplain how you incorporate physical constraints or domain knowledge into machine learning models for Earth system science. Discuss the trade-offs between purely data-driven approaches and KGML in the context of land-atmosphere feedback diagnosis.

High-Performance Computing (HPC) Workflows

What to coverDetail your experience managing large-scale climate model datasets on HPC clusters. Explain how you utilize scripting tools like Bash, CDO, or NCO to optimize data processing pipelines and ensure reproducibility.

Land-Atmosphere Feedback Mechanisms

What to coverHow would you synthesize model simulations and observational data to diagnose atmospheric responses to land surface property changes? Discuss the key variables and physical processes you would prioritize in your analysis.

Causal Inference in Environmental Modeling

What to coverDescribe how you apply causal inference techniques to distinguish between correlation and causation in complex land-atmosphere interactions. Suggest how these methods can improve the reliability of climate projections.

Bayesian Neural Networks for Uncertainty Quantification

What to coverExplain the role of Bayesian neural networks in evaluating simulation results, such as climate extremes. Discuss how these models help in quantifying uncertainty in predictions compared to standard deep learning architectures.

Programming Proficiency and Tooling

What to coverThe role requires Python, R, NCL, and MATLAB. Describe a project where you had to integrate multiple languages to achieve a research objective, highlighting how you managed dependencies and data interoperability.

Behavioural practice

Mentoring and Research Leadership

What to coverDescribe a situation where you mentored a student or research assistant on a complex technical task. How did you balance providing guidance with fostering their independent problem-solving skills?

Collaborative Research and Teamwork

What to coverProvide an example of a successful collaboration within a diverse research group. How did you contribute to the team's goals, and how did you handle any disagreements regarding methodology or data interpretation?

Communicating Complex Research to Stakeholders

What to coverDescribe a time you had to explain technical research findings to a non-expert audience or stakeholders. What strategies did you use to ensure the core message was clear while maintaining scientific accuracy?

Managing Research Deadlines and Priorities

What to coverResearch projects often involve shifting priorities and technical hurdles. Tell us about a time you faced a significant setback in your research; how did you adjust your approach to meet project objectives?

Sharpen your knowledge

Topics to prepare

Land-Atmosphere Feedback Mechanisms
Review the physical processes governing interactions between land surface properties and atmospheric responses. Exercise: Summarize how changes in surface albedo or soil moisture influence local precipitation patterns in tropical climates to prepare for diagnosing simulation results.
Knowledge-Guided Machine Learning (KGML)
Study the integration of physical constraints into machine learning models. Exercise: Research how to incorporate mass or energy balance equations into a neural network architecture to ensure predictions remain physically consistent with Earth system models.
Causal Inference in Climate Science
Understand methods to distinguish correlation from causation in observational climate datasets. Exercise: Practice applying Granger causality or structural causal models to identify drivers of ecosystem disturbance from multi-source remote sensing time series.

Skills to revise

Scientific Computing with Python
Develop workflows for processing large-scale climate datasets. Exercise: Create a script using Xarray and Dask to perform parallelized analysis on multi-dimensional NetCDF climate model outputs, simulating the requirements for high-performance computing environments.
Remote Sensing Data Processing
Gain proficiency in handling optical, radar, and lidar data. Exercise: Build a pipeline to harmonize and validate satellite imagery with ground-based sensor data, focusing on correcting for atmospheric interference and spatial resolution mismatches.
HPC Scripting (Bash, CDO, NCO)
Master command-line tools for climate data manipulation. Exercise: Write a Bash script utilizing CDO or NCO operators to perform complex operations like regridding, temporal averaging, and subsetting on large climate model simulation files.

Your preparation checklist

Synthesize your research portfolio
Prepare a summary of your published work in high-quality journals, specifically highlighting projects that utilized climate model datasets, remote sensing imagery, or Earth system modeling to demonstrate your analytical capabilities.
Demonstrate technical proficiency in HPC and scripting
Create a brief technical document or code repository samples showcasing your experience with Bash, CDO, and NCO for processing large-scale climate datasets within high-performance computing environments.
Prepare a machine learning workflow walkthrough
Develop a clear explanation of a past project where you applied machine or deep learning to environmental data, specifically addressing how you handled data harmonization or interpreted ecosystem disturbances.
Draft responses on tropical ecosystem dynamics
Formulate concise, evidence-based answers regarding your experience with tropical climate studies and your familiarity with multi-platform remote sensing data, including optical, radar, and lidar sensors.
Formulate questions for the Yanyan Cheng research group
Prepare thoughtful questions about the current NRF-funded project's specific goals, the team's approach to land-atmosphere feedback diagnosis, and the expectations for mentoring students and research assistants.

Availability last checked: 18 Oct 2026. Status can change after a check; confirm on the employer’s website. Report an issue · Our checking policy

KEEP EXPLORING

Similar AI jobs

Related skills and specializations in Singapore. Matched to this role, not your profile.

Explore more jobs

Finding similar opportunities…