Md Saifuzzaman, PhD

Department of Biology, Faculty of Science, McGill University

Dr. Saifuzzaman (Saif) is a Research Associate in the Department of Biology at McGill University, where he develops and applies monitoring and modelling tools to assess ecosystem services in the Hudson–James Bay Lowlands (HJBL) as part of a scalable framework for the Canadian Biodiversity Observation Network (CAN BON). His current research focuses on resource assessment and biodiversity monitoring using Earth observation data, big data analytics, GeoAI, and deep learning approaches.

He earned a PhD in Bioresource Engineering (Geoscience and Environmental Science) from McGill University, where his research focused on applying machine learning and artificial intelligence to model crop growth and soil dynamics through the integration of proximal soil sensing and optical remote sensing data. Prior to joining the Department of Biology, he completed a postdoctoral fellowship in Bioresource Engineering at McGill University. Earlier in his academic career, he served as a faculty member in the Department of Geography and Environment at Jahangirnagar University, Bangladesh, where he taught and conducted research in Physical Geography and GISciences. He also holds an M.Sc. from Jahangirnagar University and a Master of Environmental Studies (MES) from Queen’s University, Canada.

Dr. Saifuzzaman has collaborated extensively with local and international organizations to develop advanced geospatial technologies, including satellite- and UAV-based systems for resource mapping, environmental monitoring, and ecosystem service assessment. He has supervised and mentored students at both Jahangirnagar University and McGill University and actively contributes to the scientific community through professional societies, editorial activities, workshop organization, and peer review. His interdisciplinary research bridges environmental science, ecology, soil science, geospatial analytics, artificial intelligence, and the broader geosciences.

RESEARCH HIGHLIGHTS

Biodiversity from Space

The Hudson–James Bay Lowlands (HJBL) are among the world’s largest peatland complexes, providing critical ecosystem services and culturally significant landscapes for the Cree Nation. Despite their ecological importance, biodiversity patterns and ecosystem dynamics remain poorly understood due to limited monitoring and fragmented data. My research integrates ecological observations, Earth observation data, biodiversity records, citizen science, and field knowledge to improve understanding of species distributions and ecosystem processes in Canadian peatlands. I am developing a GeoAI-enabled monitoring framework that combines optical and microwave remote sensing with biodiversity data to derive ecosystem structure and function indicators, assess ecosystem condition, and track ecosystem-state transitions for operational biodiversity monitoring and ecosystem service assessment, supporting Canadian northern ecosystem monitoring priorities and global biodiversity targets.🌿🌎

Ecosystem Processes and Carbon Uptake using remote sensing and climate data

This research examines the spatiotemporal dynamics of carbon, water, and energy fluxes to better understand ecosystem processes. By integrating remote sensing observations and climate data, it evaluates ecosystem productivity, evapotranspiration, and radiation balance, revealing how climate variability influences ecosystem functioning across landscapes. 🌍🛰️🌱

Soil Sensing and Terrain Analysis for Field Variability

This research integrates proximal soil sensing and high-resolution topographic derivatives to characterize spatial variability in agricultural fields. Apparent soil electrical conductivity data combined with terrain metrics (e.g., slope and topographic wetness index) are used in predictive models to estimate key soil properties, including organic matter, phosphorus, and cation exchange capacity, improving soil assessment and management of agricultural ecosystems. 🌱🛰️

Probabilistic Models for Vegetation Winter Persistence

This research develops probabilistic models to assess vegetation winter persistence by integrating weather conditions, soil characteristics, and management factors. Using field observations and environmental data, the approach evaluates risks of winter damage and variability in plant survival, supporting improved monitoring and management decisions for resilient agricultural and ecological systems. 🌱📊