SI on economic costs of invasions-Feeling the pinch: global economic costs of cray sh invasions and comparison with other aquatic crustaceans

Antonin Kouba University of South Bohemia in České Budějovice, Faculty of Fisheries and Protection of Waters, South Bohemian Research Center of Aquaculture and Biodiversity of Hydrocenoses, Zátiší 728/II, 389 25 Vodňany, Czech Republic Francisco J. O cialdegui Doñana Biological Station (EBD-CSIC), Department of Wetland Ecology, Seville, Spain Ross N. Cuthbert GEOMAR Helmholtz‐Zentrum für Ozeanforschung Kiel, Kiel, Germany Melina Kourantidou Marine Policy Center, Woods Hole Oceanographic Institution, Woods Hole, MA, USA Josie South South African Institute for Aquatic Biodiversity (SAIAB), Centre for Invasion Biology, Makhanda, South Africa Elena Tricarico University of Florence, Department of Biology, Sesto Fiorentino (FI), Italy Boris Leroy Muséum National d'Histoire Naturelle, Sorbonne Universités, Université de Caen Normandie, Université des Antilles, CNRS, IRD, Unité Biologie des Organismes et Ecosystèmes Aquatiques (BOREA UMR 7208), Paris, France Rudolphe Gozlan ISEM, Université Montpellier, CNRS, EPHE, IRD, Montpellier, France Franck Courchamp Université Paris-Saclay, CNRS, AgroParisTech, Ecologie Systématique Evolution, Orsay, France Phillip J Haubrock (  Phillip.haubrock@senckenberg.de ) Senckenberg Research Institute and Natural History Museum Frankfurt, Department of River Ecology and Conservation, Gelnhausen, Germany https://orcid.org/0000-0003-2154-4341


Introduction
Owing to their sensitivity to the effects of climate change (Woodward et  partly succeeded in raising awareness of IAS-induced costs (Hoffmann and Broadhurst 2016). The existence of information on costs inferred from IAS is of utmost importance; a lack thereof may be misleading for policy making and resource management as well as minimize the awareness and preoccupation regarding IAS. Despite recent efforts to analyze invasion costs of speci c taxonomic groups (Bradshaw et  . Introductions of North American cray sh species are particularly problematic, as they often also vector Aphanomyces astaci Schikora (Oomycetes), the causative agent of cray sh plague. This oomycete is included among the IUCN 100 world's worst IAS list (Lowe et al. 2000), given the highly susceptibility and mortality of cray sh species not originating from this continent (Svoboda et al. 2017).
Invasive cray sh species (ICS) are, however, not the only invasive crustaceans with proven impacts on recipient communities (Lodge et al. 2012;Twardochleb et al. 2013). Similar to them, numerous invasive crabs have been recognized as a large threat, with marked adverse ecological and socioeconomic effects.
Examples include the Chinese mitten crab Eriocheir sinensis Milne-Edwards, 1853 and the European green crab Carcinus maenas (Linnaeus, 1758), both listed in the Global Invasive Species Database and among the 100 worst invasive species (Lowe et al. 2000). Amidst their relatively well-established presence in Europe and North America, invasive crabs are also known for their often devastating effects on the invaded environment and native biota (Holdich et al. 1999 Despite the recent advances in invasion science con rming the ecological notoriety of ICS as well as other aquatic crustaceans, economic analyses lag behind, and are partly overshadowed by the bene ts brought by aquaculture and shing industries. Direct and indirect costs associated with damages or losses from these taxa therefore remain scarce, resulting in minimal investments into research and management measures. To address this lack of information and highlight existing knowledge gaps in costs of their invasions worldwide, we utilized the InvaCost database, the most recent, comprehensive database of globally reported economic costs of IAS (Diagne et al. 2020a). This database contains detailed information on the costs (e.g. cost types, impacted sectors, regional attributes, cost estimation reliability, etc.) associated with ~ 500 invasive species. In the present study, we use a relevant subset of this database to describe the global costs associated with ICS and other aquatic invasive crustaceans to infer comparisons and understand the magnitude of their impacts. Moreover, we investigate how these costs are structured across space, time, cost types and sectors affected, as well as identify knowledge gaps in cost estimates.

Data collection and ltering
For the purpose of analyzing global costs of invasive aquatic cray sh, we used data from the InvaCost database (Diagne et al. 2020a), which primarily presents costs from English-written sources. InvaCost is a dynamic database that allows for corrections and additions of new cost entries as they develop or are reported throughout time. The rst version of InvaCost comprised 2,419 reported economic costs of IAS retrieved from published peer-reviewed and grey literature (InvaCost v1; as of December 2019). More recently, those data have been supplemented with a search for costs recorded in fteen of the most widely spoken languages, either as a mother tongue or second language (5,212 cost entries; Angulo et al. 2020), as well as via additional searches (2,374 entries; Ballesteros-Mejia et al. 2020). As of the timing of the writing (November 2020), the latest version of InvaCost (version 3.0, Diagne et al. 2020b; openly available at https://doi.org/10.6084/m9. gshare.12668570) consisted of 9,823 cost entries from IAS globally, after resolving duplications, allowing for comprehensive analysis of IAS at different taxonomic, spatial and temporal levels.
In compiling these data, grey and peer-reviewed literature were retrieved from standardised searches in online repositories (Web of Science, Google Scholar and Google search engine). The standardized searches, described in more detail in Diagne et al. (2020a), were enriched by targeted searches aiming at opportunistic collections of material containing cost information on IAS; these searches were performed through national databases, web pages of national institutions, NGOs and other organizations, as well as through contacts with regional national experts ). The collected material was thoroughly examined to assess relevance, and then scrutinized for collating cost estimates associated with aquatic crustaceans. Every cost entry recorded was described by various descriptors (Supplementary Material 1).
We identi ed cost entries attributed to invasive freshwater cray sh, based on the "Order'' classi cation by ltering out species belonging to "Decapoda" and thereafter entries belonging to relevant cray sh families ("Astacidae", "Cambaridae", "Parastacidae", and "Cambaroididae"). This resulted in a total of 112 entries. Additionally, to put the costs of invasive cray sh into perspective relative to other invasive crustaceans, we compared them to those of invasive amphipods (Order: Amphipoda; n = 1 species), crabs (Infraorder: Brachyura and Anomura; n = 6), and lobsters (Family: Nephropidae; n = 1), on the basis of reported costs in the InvaCost database. Costs for these groups were extracted using (a) the "Order" column and selecting "Amphipoda", and (b) the "Family" column and selecting families of crabs and the family "Nephropidae", respectively. The identi ed cost entries for all crustaceans thus amounted to 120 entries attributed exclusively to aquatic species. . This process allowed us to analyze in a systematic manner the total cumulative cost along the de ned period, resulting in 277 cost entries for invasive cray sh species, and considerably less for invasive crab species (n = 71), amphipods (n = 6), and lobsters (n = 2). We provided our nal dataset used as a supplement (Supplementary Material 2). (i) Method reliability: illustrating the perceived reliability of cost estimates based on the type of publication and method of estimation ("High" if costs were described in pre-assessed material (peerreviewed articles and o cial reports) or in grey material but with documented, repeatable and traceable methods; and "Low" otherwise); (ii) Implementation: referring to whether the cost estimate was actually realised in the invaded habitat ("Observed") or whether it was extrapolated based on expectations on costs beyond the invaded habitat and/or predicted over time ("Potential"); (iii) Geographic region: describing the continental geographic origin of the listed cost; costs that were not attributable to speci c regions were categorised as "Diverse/Unspeci ed"; (iv) Type of cost merged: grouping of costs according to the categories: (a) "Damage-Loss", referring to damages or losses incurred by invasion (i.e. costs for damage repair, resource losses), (b) "Management costs", comprising expenditure such as monitoring, prevention, control, eradication and (c) "Mixed" costs, including a mix of categories (a) and (b) which include cases where reported costs were not easily distinguished between damage and control costs; and (v) Impacted sector: the activity, societal or market sector that was impacted by the cost. Individual cost entries not allocated to a single sector were classi ed under "Mixed" in the "Impacted sector" column.
For the purposes of analysing the economic costs of invasive cray sh and describing trends through time, we used the calculateRawAvgCosts function implemented in the R package "invacost" (Leroy et al. in prep; borisleroy.com/invacost/Readme.html). Using this method, we calculated the observed cumulative and average annual costs between the rst recorded costs (2000) and last reported costs (2020), considering 5-year intervals.
Last we compared the costs of cray sh invasions to costs of other prominent crustacean invaders (crabs, amphipods, and lobsters) which has helped identify knowledge gaps and biases. Speci cally, we focused on comparing how the total costs of these groups varied across species, impacted geographic regions, sectors of the economy, and the type of costs.

Results
Economic costs across taxonomic groups

Economic costs based on method reliability and implementation types
Highly reliable cost entries comprised 99.8% of the documented total cost for freshwater cray sh (as well as 275 of 277 database entries, Fig. 1). Although most entries and cost estimates were classi ed as 'High' reliability, the vast majority of them were 'Potential' (US$ 1.14 billion; 89.1%) rather than 'Observed' (10.9% of total costs, US$ 141 million), implying that most were projected in time and/or space but have not necessarily been borne in practice. Note though that observed costs constituted the majority of database entries (207 out of 277 database entries, Fig. 1).
All invasive cray sh species with recorded costs in InvaCost were native to North America (Fig. 2). The majority of total ('Observed' and 'Potential') reported costs (US$ 1.23 billion; n = 232) were inferred in Europe, while US$ 54.8 million (29 database entries) was related to certain parts of North America (speci cally Wisconsin, which is north to the native range of rusty cray sh responsible for these costs) and relatively little in Asia (US$ 292.73 thousand; n = 16) (Fig. 3a). Accordingly, there was a striking absence of cost information for certain regions which include South America, Africa, and Oceania (Figs. 2 and 3).
In Europe, the vast majority of total costs were incurred in Sweden (US$ 1.03 billion; n = 24), followed by  (Fig. 3b).
Temporal dynamics of costs For invasive cray sh, the recorded total cost of US$ 1.28 billion between 2000 and 2020 ( Fig. 5a) amounted to an average annual cost over the entire period of US$ 61.14 million and to US$ 6.71 million when only observed costs were considered. Because the effects of time lags in cost reporting were not incorporated into the analyses, average cost estimates tended to reach a plateau phase in recent years (Fig. 5). Nonetheless, reported costs have increased in the last two decades by two orders of magnitude.
Costs of other crustaceans Overall, six expanded costs were inferred to amphipods, speci cally the killer shrimp D. villosus, summing up to US$ 178.8 thousand (n = 6). These costs were classi ed as "Damages and losses", impacting "Authorities-Stakeholders" solely in Europe (Italy). Lastly, the two recorded costs inferred to marine lobsters (Nephropidae) summed up to US$ 44.6 thousand. Similar to amphipod costs, costs inferred by Nephropidae predominantly impacted Authorities-stakeholders in Europe (UK), but were attributed to "Management costs" (Fig. 6b).

Discussion
The present study is the rst systematic analysis of global economic costs of invasive cray sh species and other aquatic crustaceans. Analysis of several cost descriptors helped identify key trends and knowledge gaps across spatial, taxonomic and temporal scales. Most reported cray sh costs were obtained from peer-reviewed literature and thus deemed "highly reliable", however, the vast majority were based on predictions or extrapolations arising from relatively few studies. As a result, there was a substantial difference between realized and predicted/expected costs of ICS. We identi ed four key costly species, P. clarkii, F. rusticus, F. limosus and P. leniusculus, with the latter representing the highest costs, while other damaging ICS were absent from the database. The analysis also includes comparison of ICS costs with other invasive crustaceans, namely crabs, amphipods and lobsters.
Spatial biases and persisting knowledge gaps Given these knowledge gaps, the presented costs in our study are mostly driven by P. leniusculus, inferred from damage-losses and control actions. The vast majority of these costs were the result of extrapolations, possibly indicating a lack of empirical reporting effort and monitoring. This bias is noteworthy and worrisome, as applied management efforts are seemingly not dedicated to several highrisk species, e.g. P. clarkii (

Costs of other aquatic crustaceans
Based on the reported costs of ICS, and considering that this taxonomic group remains largely understudied, it is not unreasonable to assume that costs for other related taxonomic groups such as invasive crabs or amphipods are also greatly underestimated. Having identi ed only ve invasive crabs and one invasive amphipod species with reported costs (plus only two entries associated with invasive lobsters) indicates that there likely remain substantial knowledge gaps.
In comparison to other invasive crustaceans, ICS and invasive crabs dominated in terms of reported costs (being several magnitudes higher), although the number of reported costs was also several magnitudes higher in ICS than for invasive crabs. Management intervention or cost reporting may be hindered in some cases due to public perceptions, stakeholder interests at odds and backlash, or lack of perceived necessity for management. Further, it is generally well-established in the literature that investment into control and management can lower damage-losses (Leung et al. 2002). In this study, total damages and resource losses were found to be an order of magnitude higher than control or management costs, suggesting the need for more proactive rather than reactive responses. However, management costs were dominant when considering empirically observed costs alone for ICS. Nonetheless, management attempts have largely failed at developing tools to hinder the spread or successfully eradicate widely established populations of invasive aquatic crustaceans (Gherardi et al. 2011;Stebbing et al. 2014;Haubrock et al. 2018

Data availability statement
All the data used in this study was made available as supplementary material.

Con icts of interest/Competing interests
No con ict of interest has to be declared.

Availability of data and material
The underlying data was provided as supplementary material.

Code availability
The R-code required has been referenced in the related sections within the methods.
Authors' contributions AK, FJO, PJH, RNC and MK led the writing. PJH analysed data. JS, ET, BL, RG provided valuable insights and further literature. FC created the database. All authors contributed to all aspects of the manuscript production.
Page 31/32 Figure 5 Temporal development of reported cumulative costs (total on the left vs. observed on the right) between 2000 -2020 of invasive cray sh. Points represent decadal means. Note that the last 5 years indicated in grey are subject to incompleteness in cost reporting and that the y-axes are shown on log10 scales that differ between subplots.