Ecology · Remote sensing · Artificial intelligence

Bringing research
into view

I bring together ecological observations, geospatial data and machine learning to understand landscapes and reveal environmental change.

Explore the research Adrian F. Meyer

FHNW × Data Scientist · Research & Teaching

FIG. 01 / 2019

The Hidden ForestSouthern Black Forest

Multispectral perspectives

Explore ↓01 — 16 Research projects50 Contributions ↗

01 / Research

Selected research

From France’s railway network to soil organisms in the laboratory: questions, methods and findings in context.

All research projects

Wild boar beside a railway line. Photograph: Adrian Meyer, September 2023; published by FHNW in 2025.

Current involvement/01

SNCF Réseau: AI to predict wildlife collisions

Which railway sections should be investigated first where wild boar and train traffic meet? For SNCF Réseau, Adrian F. Meyer combines satellite time series, ecological habitat modeling and a decade of collision records. The work produces annually updated habitat maps and seasonal risk rankings across the French rail network. A published habitat study and a collision-modeling preprint provide concrete findings. Camera traps, fence-mapping work and the SANGLIA research prototype connect these national models with investigation of local conditions.

Adrian F. Meyer

Denis Jordan

Theo Reibel Bhardwaj, M.Kneubühler, M.Morelle, K.Pretali, S.Pascal Baran Maxime Gombart Flavien Viguier

FHNW

SNCF

Wildlife ecology Machine learning Remote sensing Mobility

Study region · France

Screenshot of the VelObserver viewer showing a street view on Steinenring in Basel, a map inset and viewer controls.

Current involvement/02

VelObserver: assessing cycling infrastructure with data

Putting everyday cycling experiences on the map.

Adrian F. Meyer

Carolin Bronowicz

Denis Jordan Alexander Erath

FHNW

Mobility Machine learning

Project partner location · Zurich – POSMO

COLLEMBOT: automated detection of springtails in laboratory images.

Current involvement/03

COLLEMBOT: automated image analysis in ecotoxicology

Validated springtail counting for ecotoxicological reproduction tests.

Adrian F. Meyer

Denis Jordan Döring, M.Feldhaar, H.Langer, M.Mair, M.Natal da Luz, T.Souza da Silva, É.Micha Wehrli van Gestel, C. A. M.van Gestel, K.van Hall, B.van Loon, S.

FHNW

Environment Computer vision

Project partner location · Dübendorf – Eawag

Land-use and land-cover classification in the context of the Swiss land-use statistics.

Past involvement/04

Pilote IA / ADELE: artificial intelligence for land-use statistics

Combining aerial imagery and satellite time series for detailed national statistics.

Adrian F. Meyer

Denis Jordan

Natalie Lack Facchinetti, C.Hochuli, S.Leuenberger, M.Milani, G.Schär, M.Schönholzer, M.

FHNW

Geospatial data Remote sensing

Study region · Neuchâtel

Research interests

Methods in context

I work as a Data Scientist at the FHNW Institute of Geomatics and teach applied artificial intelligence. My focus is predictive environmental modelling: I develop methods that bring satellite and aerial imagery, geospatial data and ecological observations together to classify landscapes, investigate wildlife habitats and assess spatial risks.

01Wildlife ecology 02Remote sensing 03Machine learning 04Mobility 05Computer vision 06Geospatial data

02 / Publications

Knowledge that travels

Papers, conference contributions, software, datasets and talks. Connected to the projects that shaped them.

All contributions

2026Journal article

Wild boar collision data and satellite computer vision refine habitat suitability mapping across France

Adrian F. Meyer Theo Reibel Morelle, K.Kneubühler, M.Denis Jordan · European Journal of Wildlife Research, 72

Wild boar occupy a wide range of landscapes in France. This study maps seasonal habitat suitability by combining GBIF occurrence records with wildlife–vehicle collision reports from national road and rail networks. Because these…

FHNW

SNCF

2026Conference contribution

Multitemporal AI for predicting wild boar collisions on French railways

Adrian F. Meyer Theo Reibel Pretali, S.Morelle, K.Kneubühler, M.Denis Jordan · 14th International Symposium on Wild Boar and Other Suids (IWBS 2026), Zagreb, Kroatien

This conference contribution presents seasonal prediction of wild-boar train collisions on the French railway network. Time-resolved environmental variables are central: vegetation, food resources and habitat use change through…

FHNW

SNCF

2026Preprint

Multimodal Machine Learning to Predict Wild Boar Collision Risk on the French Railway Network

Adrian F. Meyer Pretali, S.Theo Reibel Bhardwaj, M.Kneubühler, M.Denis Jordan · ARPHA Preprints

This preprint addresses the spatial and seasonal prioritisation of wild-boar train collisions across the French railway network. Its prediction framework combines recorded collisions with satellite observations, landscape…

FHNW

SNCF

2026Journal article

COLLEMBOT: AI-based counting of Collembola for OECD 232 tests

Micha Wehrli Adrian F. Meyer Souza da Silva, É.van Loon, S.van Hall, B.van Gestel, C. A. M.Natal da Luz, T.Döring, M.Feldhaar, H.Mair, M.Denis Jordan Langer, M. · Environmental Toxicology and Chemistry

The OECD 232 reproduction test requires counts of adult and juvenile springtails in laboratory images. Manual counting is time-consuming and can vary between observers. COLLEMBOT automates this part of the established protocol…

FHNW

2026Software

COLLEMBOT – Collembola Counting Toolkit

Adrian F. Meyer Micha Wehrli · GitHub / Python-Software

FHNW

Science × Perception

The Hidden Forest

A different view of the landscape. Multispectral remote sensing as a scientific tool and a visual experience.

The story behind the image The Hidden Forest · Adrian Meyer · 2019

03 / Field notes

Notes from field to research

Show all

Research in practice / 2025

From field observations to risk maps

The SNCF project connects ecological observations with spatial modelling. An FHNW article explores the work behind the maps.

Science in images / 2019

A forest in invisible colours

The Hidden Forest reveals properties of vegetation that normally remain invisible to the human eye. The image received an SNSF distinction in 2019.

Teaching and continuing education / 2025

Teaching AI through practical geospatial work

In the CAS Spatial Data Analytics, machine learning and deep learning become tangible through code examples and applications in geomatics.

Fieldwork · Spectrometry / December 2018

How do you measure the spectral signature of a wild animal?

A field spectrometer, a curious deer and a question: can reflected light distinguish brown fur from the forest floor and vegetation?

Adrian Meyer holding a fixed-wing drone in a meadow with mountains in the background.

Research in the field

Adrian F. Meyer

Curiosity with a spatial perspective

My research connects animal biology and geomatics with machine learning, computer vision and scientific software development. Working with public authorities, infrastructure operators and research partners, I develop automated analyses for environmental monitoring, mobility and public geospatial datasets. Alongside this work, I am pursuing a PhD at the University of Zurich’s Department of Geography in collaboration with the FHNW Institute of Geomatics.

About me

Bringing researchinto view

Selected research

Methods in context

Knowledge that travels

The Hidden Forest

Notes from field to research

Curiosity with a spatial perspective

Bringing research
into view