The usage of a zooplankton digitization software to study plankton dynamics in freshwater fisheries
Introduction
Data on the quality and quantity of zooplankton as well as their availability and associated food preferences of fish are important parameters within fisheries science (Wagler, 1927, Nisson, 1960, O’Brien, 1979). In turn, fish predation affects zooplankton and may subsequently influence reproductive behavior as well as morphological development in size and shape (Stibor and Lüning, 1994).
For commercial fisheries in temperate regions, planktivorous whitefish play an important economic role (Anneville et al., 2009). In the last decades however, reoligotrophication of many lakes led to declined plankton abundance and subsequent reductions in whitefish stocks and growth (Klein, 1988, Valkeajarvi et al., 2002, Lumb et al., 2007). Especially changes in nutrient supply and nutrient ratios (bottom up effects) are known to have quantitative and qualitative influence on phytoplankton and thereby also on zooplankton abundance and species composition (Straile, 2000, Jeppesen et al., 2002, Müller et al., 2007). Nevertheless top down effects of fish on the lower pelagic food web (phytoplankton, zooplankton) are strongly determined by the mortality risk of individual zooplankton as an available food source and often characterized by traits related to visibility such as body size or the size of egg clutches etc. (Northcote, 1988, Hülsmann and Mehner, 1997, Eckmann et al., 2002, Stich and Maier, 2006). Hence, zooplankton plays a central role as a link between fish and primary production from both, bottom up and top down perspectives (Huntley et al., 1995). Analysis of the zooplankton diet of fish further needs data on zooplankton ingested. Therefore, stomach analyses of fish are a common and well-established method for estimating ingestion and diet preferences of fish (O’Brien, 1979, Hyslop, 1980, Müller et al., 2007).
Analyses of zooplankton samples are time-consuming, require taxonomic expertise and thereby limit sample numbers (Davis et al., 2004, Remsen et al., 2004, Benfield et al., 2007). The same applies to stomach content analyses of fish. Futher, comparability of data series is sometimes difficult due to different subjective taxonomic evaluation methods (O’Brien et al., 2017). Examining larger samples usually requires a larger number of people, which can lead to a higher probability of taxonomic and quantification errors. Another challenge arises when analyzing stomach contents to determine food preferences. The different digestibility of individual taxa and the determination of fragments result in higher identification difficulties (Elliott and Persson, 1978; Elliott, 1991) and can lead to even larger subjective differences. Automation in data processing and software-supported analysis of long-term data and their constant availability can offer new possibilities. In marine research, the Zooscan (Gorsky et al., 2010) is already an established alternative and supplemental method to traditional microscopic plankton studies. The Zooscan was developed for identification, sizing and counting of zooplankton samples. With the Zooscan, objects can be digitized and further processed by using available image analysis software. Studies using this method mainly generated data on marine zooplankton species distribution, their abundance and biomass and long-term population dynamics (Grosjean et al., 2004, Schultes and Lopes, 2009, Gorsky et al., 2010, Gorsky et al., 2019, Vandromme et al., 2012, Benedetti et al., 2019, Lombard et al., 2019, Mortelmans et al., 2019). Additionally, the Zooscan has been also used for identification of fish eggs (Lelièvre et al., 2012).
For freshwater zooplankton analyses the often large size variability of zooplankton species in freshwater and the condition of ingested food in the stomach of fish poses a challenge for automated optical validation. Additionally, egg-bearing cladocerans and copepods, which are uncommon in marine environments, are important indicators of reproductive periods within the growing season. Hence, a zooplankton imaging solution useable for freshwater fisheries purposes would allow for all above mentioned aspects.
In the present work, zooplankton samples from both a large number of whitefish stomachs and open water samples, are evaluated in terms of time effort and taxonomic resolution using Zooscan instrumentation. Here, we show how the Zooscan in combination with semi-automatic taxonomic classification can be successfully applied in fisheries specific freshwater research.
Section snippets
Field sampling
Between March and August 2017, zooplankton dynamics and food preferences of whitefish were studied biweekly using gillnet fishing and vertical zooplankton profiles in Lake Starnberg, a large pre-alpine oligotrophic lake in southern Germany.
Abundance
The analysis of the open water samples (n = 61) provided an insight into the seasonal abundance and size variability of the zooplankton community. The six vertical zooplankton abundance profiles, which are included in the analysis for each sampling date ensured a depth integrated representation of the food supply. The median zooplankton density per sampling date was between 2.2 and 8.4 ind./L. A positive trend in calanoid copepod and Daphnia spp. abundance can be seen from March to August. In
Discussion
In our study we highlight the accuracy of correct identification of different freshwater zooplankton groups in comparison to machine generated automated assignments. Additionally, we show for the first time how well the Zooscan performs in processing zooplankton data from fish stomach samples and the assessment of zooplankton reproductive periods.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
CRediT authorship contribution statement
Christian Vogelmann: Conceptualization, Methodology, Formal analysis, Investigation, Writing – original draft, Writing – review & editing. Maxim Teichert: Methodology, Formal analysis, Visualization, Writing – original draft, Writing – review & editing. Andreas Martens: Conceptualization, Writing – original draft. Michael Schubert: Conceptualization, Investigation, Writing – Original Draft. Sabine Schultes: Conceptualization, Writing – original draft. Herwig Stibor: Conceptualization,
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References (46)
- et al.
Comparison of lake whitefish (Coregonus clupeaformis)
Growth, Cond., Energy Density Lakes Erie Ont. J. Gt. Lakes Res
(2007) - et al.
Surface zooplankton size and taxonomic composition in Bowdoin Fjord, north-western Greenland: a comparison of ZooScan, OPC and microscopic analyses
Polar Sci.
(2019) - et al.
What you see is not what you catch: a comparison of concurrently collected net, optical plankton counter, and shadowed image particle profiling evaluation recorder data from the northeast Gulf of Mexico
Deep Sea Res. Part Oceanogr. Res. Pap.
(2004) - et al.
Enumeration of prey items in stomachs of European whitefish (Coregonus lavaretus L.) which contain digested fragments
Limnologica
(2006) - et al.
Assessing biases in computing size spectra of automatically classified zooplankton from imaging systems: a case study with the ZooScan integrated system
Methods Oceano
(2012) - et al.
Image Processing with Image
J. Biophoton. Int.
(2004) - et al.
Influences of human activity and climate on the stock-recruitment dynamics of whitefish, Coregonus lavaretus, in Lake Geneva
Fish. Manag. Ecol.
(2009) - et al.
The seasonal and inter-annual fluctuations of plankton abundance and community structure in a North Atlantic Marine protected area
Front. Mar. Sci.
(2019) - et al.
RAPID: research on automated plankton identification
Oceanography
(2007) - et al.
Real-time observation of taxa-specific plankton distributions: an optical sampling method
Mar. Ecol. Prog. Ser.
(2004)