I have worked on various project over my early career, spanning from early projects on feeding behaviour of red kites (Milvus milvus), a bird of prey, in Germany over smaller landscape planning projects in New Zealand using geographical information systems (GIS). My more recent projects started probably with working under Prof. Falk Huettmann at the University of Alaska Fairbanks where I started to develop my interest in machine learning algorithms and applying them to ecological questions. I started working on predicting mossy habitats on Haida Gwaii, BC, Canada which are invaluable for marbled murrelets (Brachyramphus marmoratus) as nesting sites. At the same time I started working with Grant Humphries on the project. We also started on working on projects that deal with recruitment of young scientists into professional societies, with a manuscript being submitted soon. Back in Alaska I furthered my interest into marine science that had been sporadic before. This is when I wrote my master’s thesis on predicting vertical and horizontal zooplankton distribution into the future. This was accomplished by analyzing a vast pan-Arctic dataset of physical oceanography parameters provided by Prof. Igor Polyakov and predicting oceanographic layers such as the mixed layer depth into the future. Using those layers as well as future data from the Canadian Earth System Model (CanESM2) RCP85 on chlorophyll concentration and similar I predicted the zooplankton features into the year 2100. For the modeling, the machine learning algorithms RandomForest and TreeNet were used. Groundbreaking algorithms developed by Leo Breiman (University of California Berkeley) and Adele Cutler (Stanford University) for RandomForest and Jerome H. Friedmann for TreeNet. I got hooked on marine science and went to Québec, Canada for a PhD in marine oceanography under Prof. Louis Fortier. Now, my prime research is on optical imaging of Arctic zooplankton with an in-situ camera system that produces a continuous stream of detailed images of planktonic organisms that I analyze then using machine learning algorithms to automate species recognition. I then look at the forcing of biotic and abiotic environmental parameters on this fine scale vertical distribution as well as the coupling between the primary and secondary production.

More on some research in the designated fields below.


  • Anthropogenic effects on flora and fauna worldwide; with a focus on the poles
  • Zooplankton under changing climate, future climate predictions
  • Optical imaging of planktonic organisms and particles in the oceans
  • Biogeography, spatial ecology and modeling, machine learning algorithms, advancements in statistics
  • Carbon sequestration, rainforest ecology, epiphytes and seabirds
  • Society and nature: How is the status of the environment linked to economical well-being?

A recent presentation I gave in Ottawa at the Arctic Change 2014 conference about the progress of my PhD

Research Projects

  • Developing automatic species recognition of zooplankton to the species level

    Developing automatic species recognition of zooplankton to the species level

    Based on detailed Lightframe Onsight Keyspecies Investigation (LOKI) plankton imagery (in-situ and without further subsampling), machine learning models are developed to automate species recognition in the Canadian Arctic

    Based on data collected during the ArcticNet 2013 cruise onboard the CCGS Amundsen, models are being developed for automatic identification down to the species level for important parts of the planktonic food web (limited by the size that can go through our system; the smallest possible being diatoms and the biggest being chaetognaths/small jellyfish). These models give us a vertical resolution of the water column of ~ 30 cm for a taxonomic tree of about 30-35 different species/species groups.

  • Pelagic biodiversity components in the Southern Ocean under climate change

    Pelagic biodiversity components in the Southern Ocean under climate change

    Chapter 9.1 of the Biogeographic Atlas of the Southern Ocean

    The new Atlas will be published soon, see here for more information:

  • Carbon stock hotspots in Nicaragua and Cost Rica

    Carbon stock hotspots in Nicaragua and Cost Rica

    An assessment of two tropical field sites where above - and belowground biomass as well as carbon content were estimated

    Sequestered carbon at La Suerte field station, Costa Rica and Ometepe field station, Nicaragua were estimated in a collaboration with the Maderas Rainforest Conservancy. More information soon. A poster will be presented at the US-IALE 2014 conference as well as a book chapter is in the final stages.