Projects About Publications Contact

Sefa Percival

Postdoctoral researcher · University of Hawaiʻi at Hilo

I develop remote-sensing methods and open-source software to study forest structure and environmental change. Combining multi-sensor satellite data and airborne lidar, I work on ways to align datasets, map vegetation and measure how forests vary and change across scales.

My research spans forest canopy structure, species distributions, invasive plants, and the effects of fire and land use. Working in Hawaiʻi, Madagascar and Samoa, I connect remote-sensing methods with ecological questions and build tools that make these analyses more accessible and reproducible.

Based in Hilo, Hawaiʻi, with work across the Pacific and Madagascar.

Selected work About me Download CV

Research & software

Selected contributions

View all projects →
Figure 2: a map of repeated fires in northwestern Madagascar, cumulative burned area over time, and fire severity by number of fires.

Percival et al. (2024), Fig. 2, PLOS ONE, CC BY 4.0. Landsat-8 imagery courtesy of the U.S. Geological Survey. Image source · CC BY 4.0

Repeated burns in a northwestern Madagascar forest. The map distinguishes areas burned once, twice or three times; the lower panels show cumulative burned area and fire severity. View the full figure.

Doctoral research and ongoing collaboration

Fire and forest change in Madagascar

Our study combined Landsat fire histories and GEDI lidar to examine forest recovery and repeated burning in Ankarafantsika National Park. The findings highlight the vulnerability of these dry forests and the importance of preventing further fires.

Read the study →
LiDAR points divided into vertical voxels beside equations for plant area density, plant area index and foliage height diversity.

PyForestScan voxel metrics. © 2024 Joseph Emile Honour Percival. Image source · MIT

Creator and maintainer

Measuring forest structure with PyForestScan

I created and maintain PyForestScan to bring established forest-structure calculations into Python. It turns lidar point clouds into measures of canopy height and vegetation distribution, with tools for large datasets and GIS-ready outputs.

Explore the documentation →
gwpcorMapper interface showing a Tokyo correlation map alongside a linked scatterplot.

gwpcorMapper contributors; basemap © CARTO and © OpenStreetMap contributors. Image source · GPL-3.0 · CARTO · OpenStreetMap

Creator and maintainer

Exploring spatial relationships with gwpcorMapper

A single correlation can hide local differences. I created and maintain gwpcorMapper to explore those differences through linked maps and scatterplots, including partial correlations that account for other measured variables.

Read the methods paper →

The full collection includes research software, spatial analysis, ecology, and cultural heritage documentation with Tiapapata Art Centre. Browse the projects.

Papers

Selected publications

All publications →

  1. 2026 · Accepted

    sam-stars: A multimodal, prompt-guided approach for individual tree crown delineation and species mapping

    Percival, J.; Lindiwe, K.; Shiroma, K.; Pascual, A.; Guerra-Hernández, J.; Ostertag, R.; Perroy, R.

    Remote Sensing in Ecology and Conservation

  2. 2025

  3. 2024

    Non fire-adapted dry forest of Northwestern Madagascar: Escalating and devastating trends revealed by Landsat timeseries and GEDI lidar data

    Percival, J. E. H.; Sato, H.; Razanaparany, T. P.; Rakotomamonjy, A. H.; Razafiarison, Z. L.; Kitajima, K.

    PLOS ONE