He Zhang | 张鹤

Welcome!

I am a mission-driven Data Scientist and environmental researcher with a passion for applying advanced data analytics to solve real-world challenges in agriculture and ecosystem restoration. My expertise lies in translating complex remote sensing data (UAV, satellite) and environmental datasets into actionable insights for soil health, carbon sequestration, and sustainable land management.

I completed my postdoctoral training at the Institute for Climate and Carbon Neutrality, The University of Hong Kong, advised by Prof. Jin Wu. I earned my PhD in Geography from the Université Catholique de Louvain (UCLouvain), Belgium, where I had the privilege of being supervised by Prof. Kristof Van Oost and Prof. Bas Van Wesemael. My doctoral research pioneered novel workflows for field-scale soil organic carbon (SOC) mapping using UAV-borne hyperspectral sensors.

My research integrates remote sensing, machine learning, and ecological modeling to understand and predict ecosystem dynamics, particularly in response to disturbances and climate change. I am committed to developing data-driven tools that support regenerative agriculture and inform effective, climate-resilient restoration strategies.

News!

1. Our paper "A multi-source remote sensing approach to identify and predict delayed succession in human-dominated tropical landscapes" has been accepted by Journal of Applied Ecology! Congratulations to all authors!


2. Our paper "Integrating both restoration and regeneration potentials into real-world forest restoration planning: A case study of Hong Kong" has been published in Journal of Environmental Management.

Contact Me

• Email: hzhang41@hku.hk
  zhanghe911024@outlook.com

Forecasting and Mitigating Delayed Forest Succession

My research in Hong Kong used 35 years of Landsat data to identify landscapes trapped in a state of "delayed succession" after disturbances like fire. I developed a machine learning framework to analyze the drivers, finding that proximity to remnant native forests, rather than just fire control, is the critical factor for successful recovery. This work provides a predictive tool to identify at-risk areas and guide targeted restoration efforts.

A Framework for Integrating Restoration & Regeneration Potentials

I developed a novel framework that moves beyond a one-size-fits-all approach to restoration. By using multi-temporal LiDAR data, my model quantifies both the potential for passive natural regeneration and the maximum potential achievable with active intervention. The gap between these two—the "restoration gain"—allows land managers to strategically allocate resources, focusing costly active restoration only where it is most needed and will have the greatest impact.

High-Resolution Soil Carbon Mapping with UAVs

During my PhD, I pioneered workflows for mapping soil organic carbon (SOC) in Belgian croplands at an unprecedented field scale. I developed a method to fuse high-resolution UAV-hyperspectral data with existing regional soil spectral libraries. This significantly reduces the need for extensive and costly ground sampling, making scalable and data-driven soil health monitoring feasible for regenerative agriculture.