1. Converting phytoplankton images into vector graphics
2. Creating phytoplankton patterns
3. Incorporating phytoplankton biomass and temperature data
4. Adding detail to phytoplankton panels
5. Mural production
Results document a decline in phytoplankton biomass from the 1970s to 1990s to 2010s (Figure 1). All decades exhibited a bimodal peak in chlorophyll a (chl a) concentration with the first peak occurring in winter and the second occurring in summer. The 1970s exhibited higher average chl a in winter than in summer. Conversely, the 1990s showed lower chl a in winter than in summer. The 2010s returned to a higher mean chl a concentration in winter than in summer. These results are reflected in the final product through different chl a peaks in the panels as well as with the text boxes added to emphasize different components of the chl a dataset (Figure 2).
Analysis of ecologically relevant phytoplankton taxa from Narragansett Bay revealed a wide range in abundance over time. This variation often masked any statistically significant differences in the taxa among the three decades although the HAB taxa, Dinophysis spp., and Tripos spp. (previously referred to as Ceratium) decreased (Table 1). In contrast, Thalassiosira nordenskioeldii and Skeletonema spp. increased (Table 1). Other taxa oscillated in abundance such as Eucampia zodiacus (Table 1). These results were illustrated in the final product by the increased presence of more E. zodiacus images in the 2010s compared to the 1970s and 1990s, as well as an overlayed microscopic image of E. zodiacus to bring the actual species to 'real-life' for the audience (Figure 2 & 3).
Taxa Name | Tipo | 1970-79 Mean ± SD (Cells L-1) | 1990-99 Mean ± SD (Cells L-1) | 2010-19 Mean ± SD (Cells L-1) | p-value |
Pseudo-nitzschia spp. | Diatom | 3701 ± 18235 | 5123 ± 24396 | 12919 ± 58632 | > 0.05 |
Thalassionema nitzschioides | Diatom | 81797 ± 245710 | 22909 ± 59246 | 62656 ± 292940 | > 0.05 |
Tripos spp. | Dinoflagellate | 1933 ± 703 | 500 ± 706 | 841 ± 353 | < 0.001 |
Eucampia zodiacus | Diatom | 27266 ± 27675 | 7500 ± 2121 | 90764 ± 181415 | > 0.05 |
Thalassiosira nordenskioeldii | Diatom | 76800 ± 150545 | 27000 ± 28284 | 362411 ± 376064 | 0.008 |
Odontella aurita | Diatom | 5571 ± 8541 | 5000 ± 2645 | 17750 ± 23485 | > 0.05 |
Chaetoceros diadema | Diatom | 103027 ± 239802 | 18000 ± 0 | 40402 ± 46128 | > 0.05 |
Skeletonema spp. | Diatom | 2457847 ± 7814228 | 1884674 ± 4888589 | 1349184 ± 3732765 | 0.003 |
Dinophysis spp. | Dinoflagellate | 5166 ± 8983 | 1978 ± 1840 | 2331 ± 2504 | < 0.001 |
Table 1: Phytoplankton counts. Mean (cells L-1) and standard deviation of phytoplankton concentrations for each taxa for each decade. Type designates whether the phytoplankton is classified as a diatom or dinoflagellate. ANOVA or t-test was performed to test for statistically significant differences in mean abundance among the three decades (ANOVA) or two (t-test) when low sample size was present in the 1990s (i.e., Tripos spp., Eucampia zodiacus, Thalassiosira nordenskioeldii, Odontella aurita, and Chaetoceros diadema). Significant p-values determined at α = 0.05 and indicated in bold.
Figure 1: Schematic of methodology. A) Convert microscopic image into a vector illustrative graphic, B) Create repetitive pattern for each decade (1970s, 1990s, 2010s), C) Use decadal chlorophyll a data to inform shapes of patterns. Fill background with blue to red color scheme to represent increasing water temperature, and D) Finalize product by adding text to inform distinct features in patterns and microscopic images of phytoplankton used to create illustrative graphics. Please click here to view a larger version of this figure.
Figure 2: Completed visualization. Finalized phytoplankton visualization made in the illustrator. Taxa include Thalassiosira nordenskioeldii, Thalassionema nitzschioides, Tripos spp., Odontella aurita, Skeletonema species complex, Chaetoceros diadema, Eucampia zodiacus, Dinophysis spp., and Pseudo-nitzschia spp. Please click here to view a larger version of this figure.
Figure 3: Completed art piece. Finalized phytoplankton visualization made in the illustrator alongside printed version for A) the 1970s, B) the 1990s, and C) the 2010s. Please click here to view a larger version of this figure.
Adobe Illustrator | Adobe | version 23.0.6 | Free alternatives include: Inkscape, GIMP, Vectr, Vectornator |
Eclipse E800 | Nikon | ECLIPSE Ni/Ci Upright Microscope | Now succeeded by Eclipse Ni-U |
Epson Large Format Printer | Epson | SCT5475SR | |
Heavy Matte Paper | Epson | S041596 | |
RStudio | Rstudio, PBC | version 2022.07.1 | Any statistical software tool will suffice |
Oceanographic time series provide an important perspective on environmental processes in ecosystems. The Narragansett Bay Long-Term Plankton Time Series (NBPTS) in Narragansett Bay, Rhode Island, USA, represents one of the longest plankton time series (1959-present) of its kind in the world and presents a unique opportunity to visualize long-term change within an aquatic ecosystem. Phytoplankton represent the base of the food web in most marine systems, including Narragansett Bay. Therefore, communicating their importance to the 2.4 billion people who live within the coastal ocean is critical. We developed a protocol with the goal of visualizing the diversity and magnitude of phytoplankton by utilizing Adobe Illustrator to convert microscopic images of phytoplankton collected from the NBPTS into vector graphics that could be conformed into repetitive visual patterns through time. Numerically abundant taxa or those that posed economic and health threats, such as the harmful algal bloom taxa, Pseudo-nitzschia spp., were selected for image conversion. Patterns of various phytoplankton images were then created based on their relative abundance for select decades of data collected (1970s, 1990s, and 2010s). Decadal patterns of phytoplankton biomass informed the outline of each decade while a background color gradient from blue to red was used to reveal a long-term temperature increase observed in Narragansett Bay. Finally, large, 96-inch by 34-inch panels were printed with repeating phytoplankton patterns to illustrate potential changes in phytoplankton abundance over time. This project enables visualization of literal shifts in phytoplankton biomass, that are typically invisible to the naked eye while leveraging real-time series data (e.g., phytoplankton biomass and abundance) within the art piece itself. It represents an approach that can be utilized for many other plankton time series for data visualization, communication, education, and outreach efforts.
Oceanographic time series provide an important perspective on environmental processes in ecosystems. The Narragansett Bay Long-Term Plankton Time Series (NBPTS) in Narragansett Bay, Rhode Island, USA, represents one of the longest plankton time series (1959-present) of its kind in the world and presents a unique opportunity to visualize long-term change within an aquatic ecosystem. Phytoplankton represent the base of the food web in most marine systems, including Narragansett Bay. Therefore, communicating their importance to the 2.4 billion people who live within the coastal ocean is critical. We developed a protocol with the goal of visualizing the diversity and magnitude of phytoplankton by utilizing Adobe Illustrator to convert microscopic images of phytoplankton collected from the NBPTS into vector graphics that could be conformed into repetitive visual patterns through time. Numerically abundant taxa or those that posed economic and health threats, such as the harmful algal bloom taxa, Pseudo-nitzschia spp., were selected for image conversion. Patterns of various phytoplankton images were then created based on their relative abundance for select decades of data collected (1970s, 1990s, and 2010s). Decadal patterns of phytoplankton biomass informed the outline of each decade while a background color gradient from blue to red was used to reveal a long-term temperature increase observed in Narragansett Bay. Finally, large, 96-inch by 34-inch panels were printed with repeating phytoplankton patterns to illustrate potential changes in phytoplankton abundance over time. This project enables visualization of literal shifts in phytoplankton biomass, that are typically invisible to the naked eye while leveraging real-time series data (e.g., phytoplankton biomass and abundance) within the art piece itself. It represents an approach that can be utilized for many other plankton time series for data visualization, communication, education, and outreach efforts.
Oceanographic time series provide an important perspective on environmental processes in ecosystems. The Narragansett Bay Long-Term Plankton Time Series (NBPTS) in Narragansett Bay, Rhode Island, USA, represents one of the longest plankton time series (1959-present) of its kind in the world and presents a unique opportunity to visualize long-term change within an aquatic ecosystem. Phytoplankton represent the base of the food web in most marine systems, including Narragansett Bay. Therefore, communicating their importance to the 2.4 billion people who live within the coastal ocean is critical. We developed a protocol with the goal of visualizing the diversity and magnitude of phytoplankton by utilizing Adobe Illustrator to convert microscopic images of phytoplankton collected from the NBPTS into vector graphics that could be conformed into repetitive visual patterns through time. Numerically abundant taxa or those that posed economic and health threats, such as the harmful algal bloom taxa, Pseudo-nitzschia spp., were selected for image conversion. Patterns of various phytoplankton images were then created based on their relative abundance for select decades of data collected (1970s, 1990s, and 2010s). Decadal patterns of phytoplankton biomass informed the outline of each decade while a background color gradient from blue to red was used to reveal a long-term temperature increase observed in Narragansett Bay. Finally, large, 96-inch by 34-inch panels were printed with repeating phytoplankton patterns to illustrate potential changes in phytoplankton abundance over time. This project enables visualization of literal shifts in phytoplankton biomass, that are typically invisible to the naked eye while leveraging real-time series data (e.g., phytoplankton biomass and abundance) within the art piece itself. It represents an approach that can be utilized for many other plankton time series for data visualization, communication, education, and outreach efforts.