First automatic size measurements for the separation of dwarf birch and tree birch pollen in MIS 6 to MIS 1 records from Northern Germany

Research output: Journal contributionsJournal articlesResearchpeer-review

Authors

  • Martin Theuerkauf
  • Elias Nehring
  • Alexander Gillert
  • Philipp Morten Bodien
  • Michael Hein
  • Brigitte Urban

During past glacial periods, the land cover of Northern Eurasia and North America repeatedly shifted between open steppe tundra and boreal/temperate forest. Tracking these changes and estimating the coverage of open versus forested vegetation in past glacial and interglacial landscapes is notoriously difficult because the characteristic dwarf birches of the tundra and the tree birches of the boreal and temperate forests produce similar pollen grains that are difficult to distinguish in the pollen record. One objective approach to separating dwarf birch pollen from tree birch pollen is to use grain size statistics. However, the required grain size measurements are time-consuming and, therefore, rarely produced. Here, we present an approach to automatic size measurement based on image recognition with convolutional neural networks and machine learning. It includes three main steps. First, the TOFSI algorithm is applied to detect and classify pollen, including birch pollen, in lake sediment samples. Second, a Resnet-18 neural network is applied to select the birch pollen suitable for measurement. Third, semantic segmentation is applied to detect the outline and the area and mean width of each detected birch pollen grain. Test applications with two pollen records from Northern Germany, one covering the Lateglacial-Early Holocene transition and the other covering the Mid to Late Pleistocene transition, show that the new technical approach is well suited to measure the area and mean width of birch pollen rapidly (>1000 per hour) and with high accuracy. Our new network-based tool facilitates more regular size measurements of birch pollen. Expanded analysis of modern birch pollen will help to better understand size variations in birch pollen between birch species and in response to environmental factors as well as differential sample preparation. Analysis of fossil samples will allow better quantification of dwarf birch versus tree birch in past environments.

Original languageEnglish
Article numbere11510
JournalEcology and Evolution
Volume14
Issue number6
Number of pages17
ISSN2045-7758
DOIs
Publication statusPublished - 01.06.2024

Bibliographical note

Publisher Copyright:
© 2024 The Authors. Ecology and Evolution published by John Wiley & Sons Ltd.

    Research areas

  • automatic pollen recognition, convolutional neural networks, dwarf birch, Holocene, machine learning, middle and upper Pleistocene, TOFSI, tree birch
  • Biology

DOI

Recently viewed

Publications

  1. Auf den Spuren von Heinrich Heine – Potentiale der virtuellen Schnitzeljagd Actionbound für den Literaturunterricht in der Sekundarstufe 1
  2. Finanzierung von Existenzgründungen mit Bankkrediten nach Umsetzung der Eigenkapitalunterlegungsvorschriften für Kreditinstitute (Basel II)
  3. Entwicklung und (Kosten-)Effektivitätsanalyse eines internetbasierten Programms für Personen mit Diabetes mellitus und depressiven Symptomen
  4. Eemian landscape response to climatic shifts and evidence for northerly Neanderthal occupation at a palaeolake margin in northern Germany
  5. Use of the concept of Bildung in the international science education literature, its potential, and implications for teaching and learning
  6. Conceptual approaches in the prevention of child overweight in Germany—the research project ‘Systematization of Conceptual Approaches’ (SCAP)
  7. It’s (not) the winning – The Special Olympics national handball teams in the trade-off between desire for sporting success and social support
  8. Plagiolepis alluaudi Emery, 1894, a globally spreading exotic ant (Hymenoptera, Formicidae) newly recorded from Tenerife (Canary Islands, Spain)
  9. Summer temperatures from the Middle Pleistocene site Schöningen 13 II, northern Germany, determined from subfossil chironomid assemblages
  10. Calibrated Passive Sampling - Multi-plot Field Measurements of NH3 Emissions with a Combination of Dynamic Tube Method and Passive Samplers
  11. Friction surfacing of aluminum alloys on Ti6Al4V - Investigation of process parameters, material deposition behavior and bonding mechanisms
  12. Developing ESD-specific professional action competence for teachers: knowledge, skills, and attitudes in implementing ESD at the school level
  13. Wie unterscheiden sich Familien in Bezug auf naturwissenschaftliche Anregungen im Vorschulalter? Familiäre Disparitäten im Elementarbereich
  14. Phenotypic Plasticity Explains Response Patterns of European Beech (Fagus sylvatica L.) Saplings to Nitrogen Fertilization and Drought Events
  15. XPS studies of magnesium surfaces after exposure to Dulbecco's modified eagle medium, Hank's buffered salt solution, and simulated body fluid