Wilson, O.
Microscopy-derived images and three-dimensional surface scans of spores and pollen grains from 23 plant taxa
https://doi.org/10.5285/da7799f6-9aaf-4eb7-900b-a6aa7e16a597
Cite this dataset as:
Wilson, O. (2026). Microscopy-derived images and three-dimensional surface scans of spores and pollen grains from 23 plant taxa. NERC EDS Environmental Information Data Centre. https://doi.org/10.5285/da7799f6-9aaf-4eb7-900b-a6aa7e16a597
Download/Access
By accessing or using this dataset, you agree to the terms of the relevant licence agreement(s). You will ensure that this dataset is cited in any publication that describes research in which the data have been used.
This dataset is available under the terms of the Open Government Licence
Bulk download options
You can access the data using scripting tools such as wget or curl. For example
wget https://catalogue.ceh.ac.uk/datastore/eidchub/da7799f6-9aaf-4eb7-900b-a6aa7e16a597/_filename --header="Authorization: Bearer YOUR_PERSONAL_ACCESS_TOKEN"
curl "https://catalogue.ceh.ac.uk/datastore/eidchub/da7799f6-9aaf-4eb7-900b-a6aa7e16a597/_filename" -L -O -H "Authorization: Bearer YOUR_PERSONAL_ACCESS_TOKEN"
This dataset contains microscopy-derived data from the 3D Pollen Project Knowledge Exchange Fellowship (2022-25). This forms part of the larger 3D Pollen Project (https://3dpollenproject.wixsite.com/main) which aims to make three-dimensional data about pollen widely accessible for education, outreach and research.
Confocal laser scanning microscopy was used to produce 3D data stacks from pollen grains of 23 taxa. Each taxon has at least one z-stack of images taken with a Zeiss Airyscan LSM980 microscope (in .tif format), metadata of the imaging parameters (.csv), and a 3D surface file of the scanned grain (.stl). The surface files were produced by thresholding the z-stacks in Fiji/ImageJ then simplifying and smoothing them in Meshlab. They are at 2,500x life size - that is, a 10-micron feature will be rendered 25mm in size if the model is not rescaled further. Most taxa also have a z-stack of light micrographs taken with the LSM980's T-PMT detector (.tif) to enable comparison with the confocal stacks.
Confocal laser scanning microscopy was used to produce 3D data stacks from pollen grains of 23 taxa. Each taxon has at least one z-stack of images taken with a Zeiss Airyscan LSM980 microscope (in .tif format), metadata of the imaging parameters (.csv), and a 3D surface file of the scanned grain (.stl). The surface files were produced by thresholding the z-stacks in Fiji/ImageJ then simplifying and smoothing them in Meshlab. They are at 2,500x life size - that is, a 10-micron feature will be rendered 25mm in size if the model is not rescaled further. Most taxa also have a z-stack of light micrographs taken with the LSM980's T-PMT detector (.tif) to enable comparison with the confocal stacks.
Publication date: 2026-11-04
22 downloads
107 views
Formats
TIFF, Comma-separated values (CSV), stl
Provenance & quality
Pollen samples:
Scanned material was taken from pre-prepared slides in the University of Reading Tropical Palaeoecology Reference Collection. For these, pollen had been taken from herbarium sheets and living material, processed with acetolysis, and mounted in silicone oil. The name given to each sample by the microscope slide label is noted in the metadata spreadsheet; the taxon's accepted name in POWO (https://powo.science.kew.org/) is also provided.
Scanning:
Imaging took place at the University of York Bioimaging Facility using a Zeiss LSM980 MP Airyscan 2 microscope with a LD LCI Plan-Apochromat 40x/1.2 Immersion Correction objective lens, with Immersol Sil 406 (refractive-index matched to silicone oil) as an immersion medium. The degree of immersion correction was adjusted for each sample in efforts to reduce the effects of refractive index mismatch. Confocal scans used the Airyscan detector in 4Y multiplex mode to balance resolution and speed. Light micrographs were produced by re-scanning the focal pollen grain using confocal mode and the T-PMT detector. Laser power and master gain settings were selected on a per-sample basis to optimise image quality. Representative pollen grains were scanned using software-recommended sampling dimensions in x and y, and oversampled in z (typically 0.1 micron z increments) to improve resolution in that axis. Scan settings for each Airyscan image are provided in the relevant metadata file (.csv). For archiving in the EIDC, image stacks were converted from .czi format to .tif in Fiji/ImageJ.
Processing:
Airyscan image stacks were opened in Fiji/ImageJ using the Bio-Formats plugin, converted from 16-bit to 8-bit format, and opened in the 3D Viewer plugin. Appropriate thresholds for transforming continuous image data to binary surface reconstructions were iteratively chosen for each scan and model. When an effective representation of the pollen grain's surface was reached, the .stl files were exported to MeshLab for further processing. The surface meshes had their face orientation corrected and were simplified using quadric edge collapse decimation. In some cases, image artefacts (such from as attached debris) were removed and resulting holes in the mesh closed. The surface meshes were smoothed using the HC Laplacian smoothing filter, then rescaled to be 2,500x larger than life size (ie, 1 micron of pollen = 2.5 mm of model, without further rescaling).
Scanned material was taken from pre-prepared slides in the University of Reading Tropical Palaeoecology Reference Collection. For these, pollen had been taken from herbarium sheets and living material, processed with acetolysis, and mounted in silicone oil. The name given to each sample by the microscope slide label is noted in the metadata spreadsheet; the taxon's accepted name in POWO (https://powo.science.kew.org/) is also provided.
Scanning:
Imaging took place at the University of York Bioimaging Facility using a Zeiss LSM980 MP Airyscan 2 microscope with a LD LCI Plan-Apochromat 40x/1.2 Immersion Correction objective lens, with Immersol Sil 406 (refractive-index matched to silicone oil) as an immersion medium. The degree of immersion correction was adjusted for each sample in efforts to reduce the effects of refractive index mismatch. Confocal scans used the Airyscan detector in 4Y multiplex mode to balance resolution and speed. Light micrographs were produced by re-scanning the focal pollen grain using confocal mode and the T-PMT detector. Laser power and master gain settings were selected on a per-sample basis to optimise image quality. Representative pollen grains were scanned using software-recommended sampling dimensions in x and y, and oversampled in z (typically 0.1 micron z increments) to improve resolution in that axis. Scan settings for each Airyscan image are provided in the relevant metadata file (.csv). For archiving in the EIDC, image stacks were converted from .czi format to .tif in Fiji/ImageJ.
Processing:
Airyscan image stacks were opened in Fiji/ImageJ using the Bio-Formats plugin, converted from 16-bit to 8-bit format, and opened in the 3D Viewer plugin. Appropriate thresholds for transforming continuous image data to binary surface reconstructions were iteratively chosen for each scan and model. When an effective representation of the pollen grain's surface was reached, the .stl files were exported to MeshLab for further processing. The surface meshes had their face orientation corrected and were simplified using quadric edge collapse decimation. In some cases, image artefacts (such from as attached debris) were removed and resulting holes in the mesh closed. The surface meshes were smoothed using the HC Laplacian smoothing filter, then rescaled to be 2,500x larger than life size (ie, 1 micron of pollen = 2.5 mm of model, without further rescaling).
Licensing and constraints
This dataset is available under the terms of the Open Government Licence
Cite this dataset as:
Wilson, O. (2026). Microscopy-derived images and three-dimensional surface scans of spores and pollen grains from 23 plant taxa. NERC EDS Environmental Information Data Centre. https://doi.org/10.5285/da7799f6-9aaf-4eb7-900b-a6aa7e16a597
Supplemental information
Correspondence/contact details
Author
Other contacts
Publisher
NERC EDS Environmental Information Data Centre
info@eidc.ac.uk
Rights holder
University of York
Custodian
NERC EDS Environmental Information Data Centre
info@eidc.ac.uk
Additional metadata
Funding
Natural Environment Research Council Award: NE/X001660/1
