Rapid assessment of Dreissena population in Lake Erie using underwater videography

Karatayev, Alexander Y.; Burlakova, Lyubov E.; Mehler, Knut; Hinchey, Elizabeth K.; Wick, Molly J.; Bąkowska, Martyna; Mrozińska, Natalia

doi:10.1007/s10750-020-04481-x

Cited by 20 publications

(8 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Risk assessment can be useful for prioritizing monitoring and early eradication of the highest risk invaders (Stasolla et al, 2021), especially in areas with recognized invasion debt, or a clear lag between establishment and spread (Galil et al, 2021). Monitoring and eradication efforts may benefit from more precise mapping of the spatial distribution of invasive species, such as underwater remote sensing approaches for detection and quantification of zebra mussels (Karatayev et al, 2021). Promising new approaches for managing AIS include gene-silencing, RNA-guided gene drives, and the use of transgenes as well as non-genetic approaches such as pheromone and sterile male techniques (Simberloff, 2021).…”

Section: Management Of Invasive Speciesmentioning

confidence: 99%

Aquatic invasive species: introduction to the Special Issue and dynamics of public interest

et al. 2021

View full text Add to dashboard Cite

Section: Management Of Invasive Speciesmentioning

confidence: 99%

Aquatic invasive species: introduction to the Special Issue and dynamics of public interest

et al. 2021

View full text Add to dashboard Cite

“…Mussel density maps are vital to assess their filter feeding impacts on ecosystems, as well as for nutrient management, and infrastructure planning (Depew et al 2021; Karatayev et al 2021). These maps are based on live mussel coverage (percentage area taken by mussels), mussel abundance (number of individual mussels per unit area) or mussel biomass (dry weight of mussels per unit area).…”

Section: Figmentioning

confidence: 99%

“…Mussel abundance is variable within short distances and from year to year due to substrate types, depths, and other environmental limitations (Mellina and Rasmussen 1994; Chase and Bailey 1999; Ozersky et al 2011; Strayer et al 2020). To maintain accuracy, mussel abundance maps must be updated frequently (Depew et al 2021; Karatayev et al 2021). In addition, maps showing the spatial extent of mussels are incomplete as they often rely on manual samples taken by scuba divers who use handheld dredges or Ponar grab samplers to collect mussels from a limited number of sites (Schloesser and Nalepa 1994; Custer and Custer 1997).…”

Section: Figmentioning

confidence: 99%

“…For instance, cameras fixed on benthic sleds provide images along a transect (Karatayev et al 2018). Cameras mounted to metal quadrat frames lowered from a boat are a suitable imaging technique for some sites (Ozersky et al 2011; Karatayev et al 2021). Ponar grabs with integrated cameras combine manual sampling with imaging (Lietz et al 2015).…”

Section: Figmentioning

confidence: 99%

“…Our method may be combined with more regularly‐sampled underwater imagery, for example by taking a transect from an AUV at fixed cruise altitude. This way, detailed maps of mussel distributions may be generated with greater spatial and temporal resolutions than existing techniques (Depew et al 2021; Karatayev et al 2021).…”

Section: Recommendationsmentioning

confidence: 99%

See 2 more Smart Citations

Predicting dreissenid mussel abundance in nearshore waters using underwater imagery and deep learning

Galloway

Brunet

Valipour

et al. 2022

Limnology & Ocean Methods

View full text Add to dashboard Cite

Accurate and cost‐effective dreissenid mussel abundance maps are vital to assess their ecological roles in aquatic systems. A deep neural network (DNN) modeling framework using semantic segmentation was developed to automatically assess the abundance distribution of two invasive mussel species: zebra and quagga. DNN models were trained on images captured in Lake Erie and Lake Ontario using an underwater color imaging technique. The accuracy of the method was assessed relative to manual laboratory counts of harvested mussels, their dry biomass, and percentage live coverage estimated from fixed‐size quadrats. Assessments performed on a test set collected from 2016 to 2018 show that DNN‐based mussel coverage predictions explain 79% of the variance in log biomass, and 71% for log abundance (N = 125). For reference, live coverage estimated by scuba divers was transformed and found to be a better predictor of biomass (93%) and abundance (91%) (N = 725), leaving room for improvement of our automated method. When identical images were presented to eight human analysts and the DNN, the agreement in live mussel coverage prediction was 85% (N = 189). Models generalize well to diverse underwater illuminations, camera orientations, and resolutions, but are adversely impacted by occluding vegetation and suspended sediment. DNN models are an efficient and accurate solution for mapping mussel abundances at a scale that was previously impossible. The method may be integrated with other studies to assess the mussels' impacts in a variety of aquatic ecosystems. Source code: https://github.com/AngusG/deep-learning-dreissenid and data http://10.5683/SP3/MZEBOJ for reproducing our method are publicly available.

show abstract

A dataset of individual wet weights of benthic macroinvertebrates

Hrycik,

Burlakova,

Karatayev

et al. 2024

Limnol Oceanogr Letters

View full text Add to dashboard Cite

Biomass estimates are crucial for modeling and understanding energy flow through ecosystems. Many modeling frameworks rely on published body weights of organisms to convert density estimates to biomass. However, published body weight data are limited to few taxa in a limited number of systems. Here we present mean individual weights for common benthic macroinvertebrates of the Laurentian Great Lakes from over 2000 benthic samples and 8 yr of data collection. We also compiled wet to dry weight conversions to facilitate data reuse for researchers interested in dry weight. We compared our benthic invertebrate weights to other lakes, demonstrating when weight measurements may be applied outside the Great Lakes. Sensitivity analyses supported the robustness of our calculations. Our dataset is applicable to food web energy flow models, calculation of secondary production, interpretation of trophic markers, and for understanding how biomass distribution varies by benthic invertebrate species in the Great Lakes.

show abstract

Rapid assessment of Dreissena population in Lake Erie using underwater videography

Cited by 20 publications

References 56 publications

Aquatic invasive species: introduction to the Special Issue and dynamics of public interest

Aquatic invasive species: introduction to the Special Issue and dynamics of public interest

Predicting dreissenid mussel abundance in nearshore waters using underwater imagery and deep learning

A dataset of individual wet weights of benthic macroinvertebrates

Contact Info

Product

Resources

About