Historical trends of aquatic invasive species introduction and establishment in Illinois, USA

Tracking the introduction and establishment of aquatic invasive species (AIS) is important for monitoring the biological and economic health of freshwater environments. The state of Illinois (USA) is a critical region for understanding the threats of AIS because it possesses the only continuous aquatic habitat connecting the Laurentian Great Lakes and Mississippi River Basins. In this study, we update a previous effort to catalogue and evaluate historical AIS records from Illinois. Our updated database shows that there are now at least 92 nonindigenous aquatic species established in Illinois and a further 51 have been recorded as introduced but not established. This is more species than reported in the earlier database, most likely due to improved access to data and a longer timeframe of analysis. Rates of introduction and establishment have continued to increase in Illinois over the past century, and we identify new groups of organisms that were not in the previous database. Current sampling efforts are not sufficient to detect the number of invaders present and additional non-native species may be present but not yet recorded. Illinois is likely to remain an important hub for the introduction and spread of invasive aquatic species with implications for freshwater ecosystems across the continent.


Introduction
The spread of aquatic invasive species (AIS) presents a threat to the biodiversity and resilience of freshwater ecosystems across the globe (Gallardo et al. 2015).After being introduced into a new habitat, AIS can outcompete native species, restructure food webs, and transform the abiotic conditions of entire ecosystems (Mills et al. 1994;Gallardo et al. 2015;Havel et al. 2015).Rates of introduction and spread of AIS have steadily increased over the last century and freshwater ecosystems have been impacted more strongly by invaders than other ecosystem types (Havel et al. 2015;Jacobs and Keller 2016).Degradation of the high biodiversity and economic services provided by freshwater systems is often substantial with large impacts for society and economies (Mills et al. 1994;Ricciardi and Rasmussen 1999;Rosaen et al. 2012;Bacher et al 2018).In the United States, the state of Illinois possesses the only continuous aquatic connection (via the Illinois River and Chicago Area Waterway System) between the Mississippi River Basin and the Laurentian Great Lakes Basin, making the state a pivotal gateway for spread of AIS (Jacobs and Keller 2016).An example of a current AIS issue with heightened concern of passing between the two Vol:.( 1234567890) basins via Illinois is invasive species of carp (aka Asian carp), which has been detected in water bodies increasingly close to the Great Lakes and carries considerable risks to the lakes' ecological health and economic wellbeing (Chapman et al. 2020).For these reasons, Illinois' waterways have been identified as an important area for monitoring and understanding trends in AIS introductions and establishments (USACE 2014).From the perspective of the invasion pathway, understanding which species are introduced and which have become fully established is an important step in anticipating and mitigating ecological destruction (Blackburn et al. 2011).
An effort to catalogue and assess historical records of AIS in Illinois was published in 2016 (Jacobs and Keller 2016).That study aggregated available non-native aquatic species sampling data from multiple academic institutions and state agencies between 1873 and 2012.Due to limitations of the time, the database from the Jacobs and Keller (2016) study is not freely accessible to the public.Further, the Jacobs and Keller (2016) study only considered species records documented through 2012.With the addition of eight years and a general increase in overall record availability via newer online databases, a large amount of new information about Illinois AIS introduction and establishment events over the past 177 years has become accessible (USGS 2020), allowing us to more accurately assess AIS trends and garner insights on the current status of invasions.
The aim of this project was to use this abundance of unanalyzed data to re-evaluate Illinois' historical AIS records and assess how newly available records shift our understanding of introductions and establishments within the state.We constructed a new database of all non-native aquatic species occurrences in the state of Illinois, covering the years 1842-2019.We determined which records were of established populations and which were of introduced species that failed to establish.Records were vetted to determine native/non-native status in Illinois.The database allowed us to quantify introductions and establishments in Illinois and see how AIS trends have changed over time.This represents an updated and more robust investigation of Illinois' AIS trends, which will serve future researchers and management teams addressing this key gateway between two major freshwater basins.

Database development
We defined Illinois to include all waterbodies and waterways within the state, including the portions of major rivers that form parts of the state's borders (Mississippi, Ohio, and Wabash) and the portion of Lake Michigan that contacts Illinois.Data collection began in January 2020 and concluded in March 2020.To be included in the database a record needed unambiguous species identification, geographic location to at least the county level, and a date of collection.Sources for records included the data from Jacobs and Keller (2016), and additional sources (e.g., academic and research institutions) were identified through online searches and contacted for records.All institutions contacted reported that they annually uploaded records to aggregated digital databases, such as the Great Lakes Invasive Network (GLIN) (http:// great lakes invas ives.org) and U.S. Geological Survey's Nonindigenous Aquatic Species (USGS NAS) (https:// nas.er.usgs.gov).Records from these databases were downloaded and compiled to form a complete database of occurrence records.Data was cleaned and vetted following the methods of Jacobs and Keller (2016).This involved searching for and removing all duplicates, verifying geographic data, and discarding records that did not identify the organism to species level.Due to the potential for misidentification (Kijewska et al. 2009;Vanhaecke et al. 2012), we did not include records of hybrid species.
The earliest record for each species was classified as introduced.To classify a species as established, we consulted records held by the U.S. Geological Survey (USGS), U.S. Department of Agriculture (USDA), U.S. Army Corps of Engineers (USACE), U.S. Fish and Wildlife Services (USFWS), and relevant literature, to verify that reproducing populations had been reported in Illinois (i.e., "established").
We defined aquatic non-native animals as those that met the U.S. Department of Agriculture's (USDA) definition for AIS: non-native species that live primarily in water rather than on land (USDA 2019a).Aquatic plants were defined using the USDA Plants' classification of obligate wetland: species that occur under natural conditions in wetlands with a > 99% probability (USDA 2019b).Illinois contains portions of four USDA wetland regions, and we Vol.: (0123456789) included species classified as obligate wetland in any of these regions (USACE 2018).We made a single exception and included Phragmites australis.This is classified as a facultative wetland species in Illinois, but because of its widespread distribution and large ecological impacts in Illinois wetlands we elected to include it (Able and Ragan 2003;Price et al. 2014).The non-native status of each plant species was verified primarily with the species ranges provided by USGS NAS and USDA Plants (USDA 2019b; USGS 2020).When a species' status could not be determined through these sources, we consulted relevant literature including the Global Biodiversity Information Facility website (GBIF 2020).

Analysis of introduction and establishment records
Three analyses were conducted to investigate how rates of introduction and establishment of non-native aquatic species in Illinois have changed over time.Each of these analyses was conducted separately on the introduction and establishment datasets.First, we calculated the number of new records during each decade of the database (1842-2019; see below) and used linear regression to test whether rates of introduction and establishment have changed.Second, we constructed accumulation curves for introduced and established species and used polynomial regression to assess how the rates of species discovery have changed over time.Finally, following Ricciardi (2006) we investigated longer-term changes in introduction and establishment rates by comparing the average annual rate of discovery over the entire time interval (long-term) to that of the most recent 30 years (short-term).
To determine whether current sampling efforts are sufficient to find all non-native aquatic species in Illinois, we plotted the number of established species detected per year against the annual number of records.Number of records per year serves as a proxy for total sampling effort across the state (Ricciardi 2006;Jacobs and Keller 2016), and this analysis allowed us to estimate how changes in sampling effort are related to the total number of established species detected.A logarithmic regression was fit to the data and the graph's asymptote can be used as an indication of how effective recent sampling levels in Illinois have been at detecting the total number of established species.All statistical tests were performed in R (R Core Team 2020) with a significance level of α = 0.05.

Sampling records
Occurrence records were collected from nine sources (Table 1).The U.S. Geological Survey's datasets provided the majority of records (74.7%), followed by Early Detection and Distribution Mapping System (18.6%), and the Great Lakes Invasive Network (3.8%).After the removal of duplicates, the final database had 120,511 records for 143 non-native species from eight taxonomic groups (Table 2).The earliest record in the database is the sedge Carex vesicaria, first recorded in 1842.The full dataset then runs to 2019 and includes 38 additional years relative to Jacobs and Keller (2016).Of the eight taxonomic groups (Table 2), non-native plant species occurred in highest numbers (n = 59 species), followed by fishes (n = 46 species), and mollusks (n = 16 species).Common carp (Cyprinus carpio) was the most  2022), which primarily sampled in the southern portion of the Mississippi River and targets fishes.This program also reports each fish as a record even when multiple fish were captured during a single effort.
Other data sources may include multiple individuals in each record.

Introduction and establishment
Of the 143 non-native aquatic species that have been recorded in Illinois, 92 are established (119,769 records) and 51 have failed to establish (742 records).Species that failed to establish have generally been recorded infrequently and are native to regions that are climatically different to Illinois.For example, water lettuce (Pistia stratiotes), a floating aquatic plant native to Florida, has been recorded in Illinois 13 times, but has not persisted through winter to become established (USGS 2020).Likewise, the American alligator (Alligator mississippiensis), native to the southern US, has been recorded on five occasions but has not become established (USGS 2020).
Our methods were slightly different to those of Jacobs and Keller (2016).First, we did not include records of hybrids because of difficulties with identification and the lack of genetic confirmation for these records.The Jacobs and Keller (2016) database included nine hybrids as introduced and established.Second, we removed seven plant species that were included in Jacobs and Keller (2016) because they are not considered to be obligate aquatic.If Jacobs and Keller (2016) had rejected hybrids and only included obligate aquatic species, they would have found 83 introduced and 44 established species, as opposed to the 99 introduced and 60 established species they reported.A final difference in methods is that we included species that have been recorded in the Illinois portion of Lake Michigan but not elsewhere in Illinois.This accounted for eight introduced species in our database, three of which have only been recorded after the Jacobs and Keller (2016) study, and two established species.
The number of established species reported per year has increased over time (Fig. 3).The largest number of established species reported in a single year was in 2014, when 57 species were recorded.We plotted the annual number of sampling records (a proxy for sampling efforts) against the annual number of established species recorded and observed a logarithmic relationship (Fig. 3).The curve becomes approximately horizonal at around 34 established species per year, which indicates that current sampling methods and level of effort are sufficient to record approximately 34 of the 92 established species in Illinois each year.

AIS records
Records were found for 120,511 occurrences of 143 non-native aquatic species in Illinois spanning the years 1842-2019.This is a 441.00% increase in number of records, an additional two groups (parasitic worms and diatoms), and a 44.40% increase in the number of species, relative to the findings of the Jacobs and Keller ( 2016) study.The large increase in records and species can be attributed to the addition of records from 2013 to 2019 and broader access to historical records facilitated by recent improvements in online data repositories.

Introductions
Cumulative introduction records increased at an exponential rate in Illinois from 1842 to 2019 (Fig. 2).This finding is consistent with the results in Jacobs and Keller (2016); however, our results are supported by a much larger pool of records for the same period and include an additional 38 years of data.The average rate of discovery for new introductions (0.80 species per year) for the full database has more than doubled to 1.90 over the past 30 years.This represents an increase to both the entire and recent introduction discovery rates compared to Jacobs and Keller (2016; 0.71 and 1.33 respectively).Our database revealed 55 species introductions not detected by the Jacobs and Keller (2016) database (for the same pre-2013 period) and 10 new species introductions post-2013.This increase could be caused by increased sampling, improved access to records, and/or by increased releases.Regardless of the cause, the data presented in this paper make clear that the release of non-native aquatic species in Illinois is higher than was previously believed.Additionally, deriving species introduction counts from aggregated data may contain sampling biases from individual data sources (Mangiante et al. 2018), further indicating the total number of AIS actually present in Illinois may be undercounted.As the range of species available through trades increases, and as the region warms under climate change, it is reasonable to expect that more of these species will become established.For these reasons, we recommend a general increase in sampling efforts throughout the region to better detect future  (i.e., 1999-2019) introduction events as early as possible, which would be key in targeting management efforts and potentially preventing establishment events.

Establishments
The rate of discovery of established species in Illinois per decade has significantly increased over time (Fig. 1).The Jacobs and Keller (2016) study did find an increase but not one that was statistically significant.Rates of establishment have more than doubled from the full dataset to the most recent thirty years (from 0.52 to 1.20 per year).This is substantially larger than that found by the Jacobs and Keller ( 2016) study (0.43 and 0.57 respectively).This change in the establishment rate indicates the number of invaders able to survive introduction and become invasive in Illinois is both greater than previously thought and increasing rapidly.This finding is further emphasized by the fact that our database uncovered 38 established species not detected by the Jacobs and Keller (2016) database (for the same pre-2013 period) and four new species establishments post-2013, which represents a 70% increase in established species detected in our database compared to Jacobs and Keller (2016).This highlights that new establishments are still occurring within Illinois.
Current sampling efforts detect an average of roughly 34 established species per year in Illinois (Fig. 3), or approximately one third of the 92 species that have been recorded as established.This indicates many taxa are not being effectively sampled and raises the possibility there may be established species that have not yet been discovered.It is also possible some species that were previously established are no longer extant in the state.For example, Juncus compressus was recorded as established in 1984 but has not been recorded in the state since 2011.The lack of recent records could potentially be a result of limited sampling, a declining population, or localized extinction; however, it would be difficult to confidently determine if localized extinction is occurring without a thorough sampling regimen for each species in question.The management and eradication of established species is often laborious and expensive (Keller et al. 2008), and once established, invasive species can have negative ecological impacts and act as inception points for further spread (Blackburn et al. 2011), making early detection a priority for preventive efforts.Our findings indicate both introduction and establishment events are occurring at higher rates than current monitoring efforts can fully capture, underscoring the need for enhanced AIS monitoring within Illinois and its surrounding waterways.Additional sampling could help to locate incipient invasions early and increase the chances that they could be eradicated or their spread controlled.We would also recommend continuous and broad AIS monitoring in Illinois, especially in waterways directly linking the Mississippi River Basin and Great Lakes Basin, which would help to detect new introduction events and determine when populations are expanding and shrinking in density and geographic area.We would additionally recommend additional sampling a be performed with consideration to potential biases, such as disproportionately higher attention given to species of greater concern (e.g., carp) and variation in regional sampling efforts, due to funding availability (Mangiante et al. 2018).Our database should be used as a starting point for identifying areas of concern where Illinois waterbodies have recently experienced introductions from previously undetected AIS-these areas should then be potentially targeted with additional monitoring and management work.In the same vein, we would recommend greater collaboration from management and research groups between states on the edge of both water basins, namely Indiana and Wisconsin, as the future spread, and subsequent damage, of AIS in the Great Lakes and Mississippi River Basins is not solely constrained to Illinois.Knowing when and where invasive populations are changing is important for helping decide where to best target AIS management and how much to invest in those efforts (Keller et al. 2008;Havel et al. 2015).
The data gathered for this study has been made publicly available on the USGS NAS database platform which is the leading source of aggregated nonnative aquatic species information for North America (USGS 2020).Our analysis shows that more nonnative aquatic species have been introduced to Illinois-and that more of these species have become established-than was previously known.Further, rates of new records of introduction and establishment are higher now than they have ever been.The addition of this data will help researchers and management teams understand Illinois' AIS historical and current trends, informing future monitoring studies and management efforts.The location of Illinois at

Table 1
Data sources and the corresponding number of records used for the creation of the AIS database.Source: often recorded species in Illinois (n = 71,536 records).Four carp species (C.carpio, Hypophthalmichthys molitrix, Ctenopharyngodon idella, and Hypophthalmichthys nobilis) account for 78.96% of all AIS records.Although carp species are widely distributed in Illinois, their high record count is partly a result of intensive sampling efforts of the USGS Long Term Resource Monitoring-Fish (Table1) program (USGS

Table 2
Taxonomic groups and the number of species in each that has been introduced and established in Illinois