Preparing Otolith Data for Age Estimation
A process note by Farnaz on how a large otolith dataset was cleaned, matched, and prepared for training.
Offener Kreis
Physical edition released
Research, Software & Ecology
I work at the intersection of marine research, citizen science, software development, and data-driven methods. My focus lies on supporting long-term ecological and biodiversity-related work: building systems, coordinating collaborations, and developing tools that help research remain transparent, usable, and connected to real-world contexts. This site documents ongoing projects, technical decisions, field experiences, and reflections that rarely fit into formal publications, but still shape how research unfolds in practice.
A process note by Farnaz on how a large otolith dataset was cleaned, matched, and prepared for training.
A short process note on how the Offener Kreis reflection card set evolved from a coherent layout into a clearer visual orientation system.
Building the workflow before the system: how I structured the early development of PelAtlas so that the next step is always visible.
Investigating how environmental data and simulation-based priors can support species classification in fisheries sonar data, with a focus on sprat and herring in the Baltic Sea.
A research project on machine learning–supported age determination of Baltic cod from otolith microscopy images. The work focuses on data consolidation, quality assessment, and the development of supervised models to support expert-based age reading.
Building a scalable, ML-assisted workflow for marine mammal photo-ID and capture–recapture studies, combining retrieval-based re-identification with human verification and active learning.
Developing a set of reflection cards and minimalist posters to support structured conversations about participation, roles, and decision-making in research projects.