Benthic invertebrates are important trophic links in food webs and useful bioindicators of environmental conditions, but long-term benthic organism abundance data across broad geographic areas are rare, and historic data sets are often not readily accessible. This data set for a companion paper by Karatayev et al. (2022) provides densities of benthic macroinvertebrates collected from 1930 to 2019 during surveys in Lake Erie, a Laurentian Great Lake. The surveys were funded by the governments of the United States and Canada to investigate the status and changes in benthic community. From the total of 21 lake- and basin-wide benthic surveys conducted in Lake Erie from 1929 to 2019, we were able to acquire data for 17 surveys, including species-level data for 10 surveys and data by higher taxonomic groups for 7 surveys. Our amassed Lake Erie dataset includes data from 11 surveys (including 5 with species-level data) conducted in the western basin in 1930–2019, 7 surveys (6 with species-level data) in the central basin, and 8 surveys (7 with species-level data) in the eastern basin (1973–2019). This Lake Erie dataset represents the most extensive temporal dataset of benthic invertebrates available for any of the Laurentian Great Lakes. Benthic samples were collected using Ponar or Shipek bottom dredges and taxa densities were calculated as individuals per m² using the area of the dredge. Density data are provided for taxa in the Annelida, Arthropoda, Mollusca, Cnidaria, Nemertea, and Platyhelminthes phyla. Current taxonomy was used for most groups, but in a few cases, older taxonomic names were used for consistency with historical data. Analysis of this dataset indicated that eutrophication, water quality improvement, and dreissenid introduction were the major drivers of changes in the benthic community in the western basin, while hypoxia was a major factor in the central basin, and dreissenid introduction was most important in the eastern basin. The previously diverse benthic community of the western basin has changed dramatically over 90 years, transitioning from a diverse benthic species assemblage indicative of good water quality in the 1930s to a one of low- diversity assemblage dominated by oligochaetes and other pollution-tolerant species in the 1960s, followed by recovery in the early 2000s to a state similar to that reported in 1930. In contrast, the central basin benthic community has over the past 60 years been consistently dominated by low oxygen-tolerant taxa, signifying the persistence of hypoxia. The eastern basin community changed dramatically in recent decades, with the disappearance of Diporeia after the introduction of Dreissena in the 1990s and more recent declines in oligochaetes, amphipods, gastropods, sphaeriid clams, and leeches. Considering the rarity of high taxonomic resolution long-term benthic data for lake ecosystems, this data set could be useful to explore broader aspects of ecological theory, including effects of eutrophication, hypoxia, invasive species, and other factors on community organization, phylogenetic and functional diversity, and spatial and temporal scales of variation in community structure. In addition, the data set could be useful for studies on individual species including abundance and distribution, species co-occurrence, and how the patterns of dominance and rarity change over space and time. Use of this data set for academic or educational purposes is encouraged. In doing so, we kindly request the data source be properly cited using the title of this dataset, the names of the authors, the year of publication, and the link to Dryad. 

Lake Erie (surface area 25,690 km²) is one of five Laurentian Great Lakes of North America centered at 42°15’N and 81°15’W. It is bounded on the north by the Canadian province of Ontario, and on the south, west, and east by the U.S.A. states of Michigan, Ohio, Pennsylvania, and New York (moving west to east). The Canadian and the U.S. water boundaries meet in the middle of the lake.

Benthic habitats where samples were collected ranged from depths 2 to 63.5 m. All benthic invertebrate data were collected during surveys conducted in the summer and fall months between 1930 and 2019. Historical basin-wide surveys in Lake Erie varied in sample design, number of stations sampled per year, sampling gear (Ekman, Petersen, Franklin-Anderson, Shipek, and Ponar grabs), the mesh size of the sieve used to wash sediments (180 – 760 µm), and the numbers of surveys per year (Karatayev et al., 2022, Table 1). The level of taxonomic identification and taxa reported also varied among surveys from reporting only Oligochaeta and Hexagenia to reporting all species collected (Karatayev et al., 2022, Table 1). The weight of benthic invertebrates was measured only in a few surveys and different methods and units were used to report biomass, including volume (wet ml/sample) in 1962 – 1965 (Barton, 1988), dry weight in 1978, (Dermott, 1994), and wet weight in 2014 and 2019 (Burlakova et al., 2017; Karatayev et al., 2021). To reconstruct the benthic wet biomass for all years except 2014 and 2019, we multiplied the density of the major taxa (Oligochaeta, Chironomidae, Sphaeriidae Dreissena polymorph and D. r. bugensis, and Hexagenia) by their individual wet weights. Individual wet weights were estimated separately for each taxon and each basin by dividing the taxon's average biomass for 2019 by their average density (Hrycik et al. in preparation). 

Western basin

A total of 18 benthic surveys were conducted in the western basin of Lake Erie between 1929 and 2019 (Karatayev et al., 2022). These studies varied in the number of stations sampled (8 – 63), number of sampling events (1 – 24 samplings), and duration (from a few days to three years).

1929 – 1930. The 1929 – 1930 studies (Wright 1955) were conducted to assess the status of benthic and water resources of western Lake Erie. The 1930 study was one of the first and most extensive benthic studies conducted in the Great Lakes in the early 20th century (Schloesser et al. 2017). In 1929, benthic invertebrates were sampled from 14 stations only, using an Ekman dredge. In 1930 a more extensive survey was conducted that sampled 80 stations using a Petersen dredge (Wright 1955). In addition, the 1929 study was less consistent as different numbers of replicate samples were obtained at individual stations and sampling periods, thus we excluded the 1929 survey from our analysis.

Table 1. Lake-wide (or basin-wide) benthic surveys conducted in Lake Erie between 1929 – 2019. Surveys marked with shading were excluded from analysis due to small sample sizes, incompatible methods, or unavailable data.

^aThe number of stations where samples were collected successfully (n indicated in parenthesis) and data are available.

^bNo primary data available.

^cThe number of additional stations where samples were collected for Dreissena only (n indicated in parenthesis).

^dNo data for Dreissena spp.

*Surveys that contain the nine consistently sampled stations.

From the 80 stations sampled in 1930, 13 stations were located in areas with hard bottom preventing the collection of quantitative samples, and therefore only 67 stations were sampled successfully (Wright 1955). Unfortunately, during the 1930 study laboratory analysis methods varied in the level of data resolution and the number of taxa reported. While Tubificidae and Hexagenia densities were reported from all 67 stations, only 11 stations contain information for all taxa, 8 additional stations also have data for Chironomidae, and Sphaeriidae, and 29 additional stations have data for Gastropoda, therefore different numbers of stations were used to calculate averages for different taxa. 

1961. A number of local studies of bottom fauna in the western basin of Lake Erie were conducted in the 1950s (Brown 1953; Britt 1955a, 1955b; Beeton 1961), but the next basin-wide survey to be conducted was not until 1961 (Carr and Hiltunen 1965; Table 1). Carr and Hiltunen (1965) collected benthic samples using a Petersen dredge from 40 stations, including revisiting 33 stations from Wright’s 1930 study. This study was designed to assess the impacts of nutrient pollution primarily from rivers that carried wastes away from the population centers of Detroit, Michigan, and Toledo, Ohio. As such, the 1961 study was systematic in that it included samples primarily from near river mouths (open water sites were also included primarily for reference benthos), collected in a short period of time in spring before most insects emerge from the water (May – June), consistently collected an equal number of replicates at each site (n = 3), and followed detailed and consistent laboratory methodology (Schloesser et al. 2017).

1963 – 1967. In 1963 – 1965, the University of Toronto Great Lakes Institute conducted an intensive lake-wide benthic survey comprised of 83 stations annually (11 stations in the western basin, 45 stations in the central basin, and 27 stations in the eastern basin) during 24 cruises (Barton 1988). This was the first and the most extensive lake-wide survey in Lake Erie. In addition, this was the first basin-wide benthic study in both central and eastern basins (Table 1). Unfortunately, this survey has several serious sampling limitations including inefficient sampling gear (use of a Franklin-Anderson grab as opposed to a Ponar or Petersen grab) and problems with the storage and laboratory processing of samples. The efficiency of a Franklin-Anderson grab is very low and often collected samples that did not contain any benthic organisms (Brinkhurst et al. 1968; Barton 1988; Barton and Anholt 1997). According to Barton (1988), densities averaged by Lake Erie basins in 1962-1965, “were only one-tenth to one-third as large as those reported from other studies, most likely due to the way the samples were handled prior to preservation”. Moreover, 20 – 30% of the samples collected from the central and western basins during June – August contained no living animals. Because the study was conducted during the peak of eutrophication in Lake Erie when extremely high oligochaete densities were known to be present, the large underestimation of benthic organisms during the 1963 – 1965 surveys is obvious. Therefore, we excluded this survey from our analysis. We were not able to find data for the 1967 survey mentioned in Veal and Osmond (1968).

1970 – 1975. In July – August of 1970, 12 stations were sampled by the University of Michigan, Ann Arbor using a Ponar grab and washed through 760 µm nylon screen (Table 1; Schelske and Roth 1973). This survey was followed by a 1973 – 1975 study of 13 stations in the western basin and 36 stations in the central basins (Britt et al. 1980). From this study, we were able to find only 1973 data collected during June – October cruises (Herdendorf 1979).

1978 – 1979. A large, lake-wide benthic survey of 158 stations was conducted in 1978 –1979 that included 52 stations in the western basin (sampled in 1979 only), 69 stations in the central basin, and 37 stations in the eastern basin (sampled in 1978 only) (Dermott 1994). Unfortunately, this study utilized a Shipek grab, and samples were washed through a 153 µm mesh, factors which limit the comparability of this survey to other surveys (Table 1).

1982 – 2019. In June of 1982, benthic samples in the western basin were collected from 40 stations (Table 1). This survey was performed to assess the progress of pollution-abatement programs that were initiated in the early 1970s. J. K. Hiltunen (U.S. Fish and Wildlife Service, Great Lakes Fishery Laboratory, Ann Arbor, MI) performed both the 1961 and 1982 studies and provided field and laboratory guidance to maintain continuity of study comparisons. The following surveys were conducted by the U.S. Geological Survey in June 1993 (47 stations sampled), April – May 2003 (60 stations), March – April 2010 (31 stations), and April 2014 (31 stations) (Schloesser et al., 2017; Kennedy et al. 2023). 

Fourteen stations were sampled in June – September of 2009 (Burlakova et al., 2014). Based on data collected in 2009 and smaller surveys undertaken in 2011 and 2012, Burlakova et al. (2014) documented significant changes in benthic community composition from 1963 to 2012, with the largest difference found between pre- and post-dreissenid invasion communities. While our study confirmed these findings, the 2009 survey was largely biased toward the shallow (<5 m) zone along the southern shore of the lake exhibiting very low benthic density, with several stations having no organisms at all (Burlakova et al., 2014). In addition, benthic organisms in this survey were live-picked, which also may contribute to underestimation of density, compared to picking organisms from preserved samples under the dissection scope as was done in other surveys. The combination of all these factors led to the unusually low benthic density (e.g., western basin average oligochaete density in 2009 was < 10% of the density measured in 2010). Therefore, we excluded the 2009 survey from the analysis.

Eighty benthic stations were sampled in July-August of 2014 by the SUNY Buffalo State Great Lakes Center (GLC) collecting three replicate samples at each station with a Ponar grab. Each sample was placed separately into an elutriation device and then washed through a 500-µm mesh screen. All retained organisms and sediments were placed into a collection jar and preserved with neutral buffered formalin with Rose Bengal stain to a final formalin concentration of 5 – 10%. Details are described in Schloesser et al. (2017) and Burlakova et al. (2017). Both April and summer 2014 samples were processed by the GLC and were combined for the long-term analysis.

Finally, Lake Erie benthos was studied in April 2019 and July-August 2019 by the EPA and GLC. In 2019, a total of 249 benthic samples were collected at 77 stations, including 55 stations sampled aboard R/V Lake Guardian in July during the Lake Erie Cooperative Science and Monitoring Initiative (CSMI, 2022) benthic survey and 10 Long-Term Monitoring (LTM) stations sampled during the U.S. Environmental Protection Agency (US EPA) Great Lakes Biology Monitoring Program (GLBMP 2022) summer survey in August using a Ponar grab (sampling area 0.0523 m²). Three shallow CSMI stations (973, DO2, ER03) in the western basin were sampled by a National Oceanic and Atmospheric Administration (NOAA) small vessel in July (Fig. 7). Another 25 CSMI stations were sampled aboard the R/V Lake Guardian in July only for the assessment of the distribution and population size of Dreissena. In addition, 9 stations in the western basin were sampled for benthos on April 24, 2019, aboard SUNY Buffalo State’s John J. Friedhoff using a petit Ponar (sampling area 0.0231 m²). These 9 stations (3D, 8D, 15D, 2L, 6L, 1M, 7M, 8M, and 4R) were sampled consistently over 90 years (see detail description below). All field operations were conducted according to the EPA Standard Operating Procedures for Benthic Invertebrate Field Sampling SOP LG406 (US EPA, 2019). Three replicate Ponar samples were collected from each station to determine benthic species richness, density (number of individuals/m²) and total wet biomass (g/m²). All samples collected in 2019 were elutriated through a 500 µm mesh sieve and preserved with neutral buffered formaldehyde with Rose Bengal stain to a final formalin concentration of 5 – 10%. Environmental parameters used in this study were collected 1 m above the bottom from each of the stations using a Rosette sampler equipped with a Seabird CTD, WETlab C‐Star transmissometer measuring beam attenuation due to particles (660 nm wavelengths) across a 25 cm pathlength, and a Sea-Bird Scientific SBE 43 DO sensor, all deployed from the R/V Lake Guardian.

Details of laboratory sample processing are described in the EPA Standard Operating Procedure for Benthic Invertebrate Laboratory Analysis SOP LG407 (US EPA, 2015) and in Karatayev et al. (2022). All benthic invertebrates were picked out of samples under low magnification using a dissecting microscope. All Dreissena were identified to species, counted, and measured using a digital caliper (0.01 mm). Next, all mussels in each replicate were combined into 5 mm size groups and weighed with shells to the nearest 0.0001 g after being blotted on absorbent paper. Other invertebrates (Amphipoda, Chironomidae, Oligochaeta, Mollusca) were identified, counted, and weighed after being blotted dry. Adult oligochaetes were identified to species; immatures were taken to the lowest taxonomic level possible, usually family, and included in abundance estimates. Oligochaete fragments, though counted, were excluded from density but used for biomass estimates because they could be weighed but not attributed to individuals. Other invertebrates were identified to species or genus, when possible. Meiobenthic organisms (e.g., Nematoda, Hydracarina, Ostracoda, benthic Cladocera, Copepoda, and Harpacticoida) were not recorded in our samples and were excluded from historical data analysis, if present. 

Data reported by Dermott and Dow (2008) from the western basin for 1992, 1993, and 1998 had fewer than 10 benthic stations per survey and were excluded from all benthos, but 1992 and 1993 data were included in the analysis of Dreissena population dynamics as more stations were sampled for Dreissena than for all benthos. Therefore, from a total of 18 studies conducted in the western basin of Lake Erie over 90 years, we were only able to use data from 11 surveys in our analysis. Nevertheless, available data represent the longest and most complete dataset of benthic invertebrate time series available for the Great Lakes.

Another unique aspect of the long-term benthic data set collected in the western basin is that while there was a large variation in the number of stations and sampling designs over 90 years, a subset of 9 stations (3D, 8D, 15D, 2L, 6L, 1M, 7M, 8M, and 4R) was sampled more consistently by the USGS Great Lakes Science Center (formerly the Bureau of Commercial Fisheries and U.S. Fish and Wildlife Service) in 1930 (Wright, 1955), 1961 (Carr and Hiltunen, 1965), 1982 (Manny and Schloesser, 1999), 1993, 2003, and 2010 (Kennedy et al. 2023), 2014 (Schloesser et al., 2017), and in 2019 by the SUNY Buffalo State Great Lakes Center (Karatayev et al. 2022). All of these surveys were conducted during springtime, with a similar protocol, using the same mesh size (500 – 600 µm), and are likely the most comparable dataset of all the surveys. Please note that the 1993 and 2010 studies conducted in on these 9 stations do not report data on Dreissena.

Central and eastern basins

A total of 10 benthic basin-wide surveys were conducted from 1963 – 2019 in both the central and eastern basins (Table 1). As indicated above, data collected by the University of Toronto Great Lakes Institute in 1963 – 1965 were excluded from the analysis. In 1973 and 1975, the central basin was repeatedly surveyed (4 – 5 times per year) by Britt et al. (1980) at 36 stations. Although we were not able to locate primary data for 1974 and 1975, data for 1973 were reported by Herdendorf (1979). In 1973 – 1976, the eastern basin was surveyed 1 – 3 times per year by Flint and Merckel (1978) at 26 stations, however, we were able to locate only averaged data for 1973 – 1975 (Mullin, 1980) and another dataset reported by species and stations averaged for all surveys conducted in 1976 (GLC, 1978 report) (Table 1). In 1992 and 1993 respectively, 22 and 20 stations were sampled in the central basin and 15, 10, and 13 stations were sampled in 1992, 1993, and 1998 (Dermott and Dow 2008; Table 1). Again, central basin surveys conducted in 1998 had fewer than 10 stations sampled (Dermott and Dow, 2008) and were excluded from the analysis of the long-term dynamics of benthos (Table 1).

Organisms were picked out of samples under low magnification using a dissecting microscope. Oligochaetes and chironomids were mounted on slides and identified under a compound microscope; other organisms were identified under a dissecting microscope. Adult oligochaetes were identified to species; immatures were taken to the lowest taxonomic level possible, usually family, and included in abundance estimates. Fragments were excluded from density. Chironomids were identified to the lowest practical taxonomic level, usually genus. Other invertebrates were identified to species, when possible. Meiobenthic organisms occasionally present in samples (e.g., Nematoda, Hydracarina, Ostracoda, benthic Cladocera, Copepoda, and Harpacticoida) were not recorded in our samples and were excluded from historical data. We report the average density of macroinvertebrate (invertebrates retained by a 500 µm sieve) taxa (individuals m^-2) found in all replicate samples collected at each station. For the 2019 survey sample, processing is described in Standard Operating Procedure for Benthic Invertebrate Laboratory Analysis (SOP LG407) (US EPA 2015). 

In 2019, taxonomy followed Kathman and Brinkhurst (1998) (Oligochaeta); Holsinger (1972) and Bousfield (1958) (Amphipoda); Epler (2001) and Wiederholm (1983) (Chironomidae); and Smith (2001), Merritt et al. (2008), and Thorp and Covitch (2001) (for other groups). Details are described in Standard Operating Procedure for Benthic Invertebrate Laboratory Analysis (SOP LG407, Revision 09, April 2015) (US EPA 2015). Due to inconsistency of the level of taxonomic resolution across surveys, species with rare occurrences (at 1 to 3 stations total in the whole data set) were pooled into higher taxonomic units (e.g., family). We used the Integrated Taxonomic Information System (ITIS, https://www.itis.gov/) for current taxonomy, but in a few cases, we used taxonomic names listed in common identification guides (i.e. species groups) or older names consistent with historical data (i.e. Sphaeriidae for family Pisidiidae; Mysis relicta; Physa).

We published an analysis of this dataset and station averages of species densities for each year in this article: Karatayev, A. Y., L. E. Burlakova, A. R. Hrycik, S. E. Daniel, K. Mehler, E. K. Hinchey, R. Dermott, R. Griffiths. 2022. Long-term dynamics of Lake Erie benthos: One lake, three distinct communities. Journal of Great Lakes Research. 48(6): 1599-1617. https://doi.org/10.1016/j.jglr.2022.09.006