Helena Filipsson
Professor
Addressing the segmentation challenge posed by 3D pore patterns and thickness of foraminiferal tests
Författare
Summary, in English
Over the past two centuries, coastal regions have experienced various anthropogenic environmental stressors such as ocean acidification, pollution, warming, and deoxygenation that have impacted benthic marine life. Understanding the severity and potential outcomes of such changes is crucial for supporting effective environmental management strategies.
Benthic foraminifera, protists with calcite tests, have long been recognized as excellent recorders of past bottom-water conditions. Morphological parameters of the shell, such as thickness and pore patterns, are of great interest for environmental reconstruction. Shell thinning is a known consequence of ocean acidification, and pore patterns are increasingly used as a proxy for bottom water oxygenation. However, the relationship between these two parameters is not well defined due to challenges posed by species-specific traits, shell curvature, and limited access to test thickness or pore blockage.
Recent advances in morphological analysis of foraminiferal tests using microcomputed tomography (μCT) have led to significant progress in generating 3D reconstruction tests. The 3D approach enables a non-destructive study of the
morphology, which is beneficial for further geochemical analyses or studying legacy museum collections. To draw statistically valid conclusions, it is necessary to scan as many tests as possible and work at sub-micrometer resolution for
measurement accuracy. Synchrotron light-based approaches can be used to achieve these objectives. However, extracting the required parameters from 3D tests remains challenging due to the limitations of image processing and computational
capacity; therefore, optimizing post-data analysis is crucial.
Previously, it was described a post-data routine for analyzing entire tests in 3D from Elphidium clavatum specimens that recorded environmental conditions in the Baltic Sea entrance from the early industrial (the 1800s) and present-day (the
2010s) conditions. The 3D time series of morphological parameters revealed that modern specimens have on average 28% thinner tests and 91% more pores than their historical counterparts. These morphological changes were interpreted as the
result of gradual environmental changes in the Baltic Sea inlet that have intensified since the start of the industrial era, linked in particular to a decline in pH and an increase in the duration and severity of hypoxia in the region.
Here, we have extended the analyses to Elphidium clavatum specimens to explore changes in their pore patterns from a selected chamber. The challenge of achieving 3D pore patterns remains, particularly concerning cropping a chamber from an
image stack and segmenting and reconstructing pores in 3D. We detail the following parameters: porosity (%), pore area (μm2), crop area (μm2), number of pores, pore density, maximum pore size, minimum pore size, standard deviation of pore
sizes, and average pore size. Thus, the main interest is therefore to continue to gain efficiency in image processing and to rapidly generate large databases of 3D morphological parameters. To accelerate the adoption of this semi-automated
approach, we developed a routine using open-source software in ImageJ and Matlab (student version). The perspectives of this work are multiple, as it would allow for a better understanding of how the morphology of the shell varies during
ontogeny or culture experiments under controlled conditions.
Benthic foraminifera, protists with calcite tests, have long been recognized as excellent recorders of past bottom-water conditions. Morphological parameters of the shell, such as thickness and pore patterns, are of great interest for environmental reconstruction. Shell thinning is a known consequence of ocean acidification, and pore patterns are increasingly used as a proxy for bottom water oxygenation. However, the relationship between these two parameters is not well defined due to challenges posed by species-specific traits, shell curvature, and limited access to test thickness or pore blockage.
Recent advances in morphological analysis of foraminiferal tests using microcomputed tomography (μCT) have led to significant progress in generating 3D reconstruction tests. The 3D approach enables a non-destructive study of the
morphology, which is beneficial for further geochemical analyses or studying legacy museum collections. To draw statistically valid conclusions, it is necessary to scan as many tests as possible and work at sub-micrometer resolution for
measurement accuracy. Synchrotron light-based approaches can be used to achieve these objectives. However, extracting the required parameters from 3D tests remains challenging due to the limitations of image processing and computational
capacity; therefore, optimizing post-data analysis is crucial.
Previously, it was described a post-data routine for analyzing entire tests in 3D from Elphidium clavatum specimens that recorded environmental conditions in the Baltic Sea entrance from the early industrial (the 1800s) and present-day (the
2010s) conditions. The 3D time series of morphological parameters revealed that modern specimens have on average 28% thinner tests and 91% more pores than their historical counterparts. These morphological changes were interpreted as the
result of gradual environmental changes in the Baltic Sea inlet that have intensified since the start of the industrial era, linked in particular to a decline in pH and an increase in the duration and severity of hypoxia in the region.
Here, we have extended the analyses to Elphidium clavatum specimens to explore changes in their pore patterns from a selected chamber. The challenge of achieving 3D pore patterns remains, particularly concerning cropping a chamber from an
image stack and segmenting and reconstructing pores in 3D. We detail the following parameters: porosity (%), pore area (μm2), crop area (μm2), number of pores, pore density, maximum pore size, minimum pore size, standard deviation of pore
sizes, and average pore size. Thus, the main interest is therefore to continue to gain efficiency in image processing and to rapidly generate large databases of 3D morphological parameters. To accelerate the adoption of this semi-automated
approach, we developed a routine using open-source software in ImageJ and Matlab (student version). The perspectives of this work are multiple, as it would allow for a better understanding of how the morphology of the shell varies during
ontogeny or culture experiments under controlled conditions.
Avdelning/ar
- Geologiska institutionen
- Statistiska institutionen
Publiceringsår
2023-06
Språk
Engelska
Sidor
74-74
Länkar
Dokumenttyp
Konferensbidrag: abstract
Ämne
- Other Earth Sciences (including Geographical Information Science)
- Probability Theory and Statistics
Conference name
International Symposium on Foraminifera
Conference date
2023-06-26 - 2023-06-30
Conference place
Perugia, Italy
Aktiv
Published