The recorded economic costs of alien invasive species in Italy

Whilst the ecological impacts of invasion by alien species have been well documented, little is known of the economic costs incurred. The impacts of invasive alien species on the economy can be wide-ranging, from management costs, to loss of crops, to infrastructure damage. However, details on these cost estimates are still lacking, particularly at national and regional scales. In this study, we use data from the first global assessment of economic costs of invasive alien species (InvaCost), where published economic cost data were systematically gathered from scientific and grey literature. We aimed to describe the economic cost of invasions in Italy, one of the most invaded countries in Europe, with an estimate of more than 3,000 alien species. The overall economic cost of invasions to Italy between 1990 and 2020 was estimated at US$ 819.76 million (EUR€ 704.78 million). This cost was highest within terrestrial habitats, with considerably fewer costs being exclusively associated with aquatic habitats and management methods, highlighting a bias within current literature. There was also a clear indication of informational gaps, with only * Contributed equally as the first authors. NeoBiota 67: 247–266 (2021) doi: 10.3897/neobiota.67.57747 https://neobiota.pensoft.net Copyright Phillip J. Haubrock et al. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. RESEARCH ARTICLE Advancing research on alien species and biological invasions A peer-reviewed open-access journal


Introduction
Despite an increasing number of indicators, targets and alarming reports on the rapid decline of biodiversity worldwide, limited economic resources have been allocated to tackle the ongoing erosion of biodiversity (Gren et al. 2009;Hulme et al. 2009;Scalera 2010). Amongst the key drivers of biodiversity decline are the ecological impacts of invasive alien species (IAS) on native species and ecosystems (Malcolm and Markham 2000;Stigall 2010), with the spread of alien species showing no sign of abatement (Seebens et al. 2017). Alien species have been shown to be major drivers of extinction globally across multiple taxonomic groups and geographic regions, being the second most common threat associated with extinct species since AD 1500 .
The ecological impacts of biological invasions have been well-described and reported in abundance in scientific literature (see Gurevich and Padilla 2004;Didham et al. 2005;Dick et al. 2017;Haubrock et al. 2021a). For example, plant invasions have been shown to have significant impacts at the species, community and ecosystem level, reducing the diversity and abundance of native assemblages (Vilà et al. 2011). In turn, impacts from groups, such as mammalian predators, have been shown to be particularly marked on native birds, mammals and reptiles, especially on islands (Doherty et al. 2016). However, relatively-few studies have synthesised impacts in monetary terms (Bradshaw et al. 2016;Diagne et al. 2020a, b), limiting economic quantifications of invasion costs. Indeed, reported economic costs of IAS are fragmented across habitats (e.g. terrestrial, aquatic), specific management actions (e.g. control, eradication) or activity sectors (e.g. agriculture, fisheries; Lovell et al. 2006;Marbuah et al. 2014). Accordingly, there is a distinct lack of comprehensive quantification regarding economic costs of invasive species across multiple systems and geographic regions. Characterising the cost of invasions at the national level, the main governance level at which budgets are established, is essential to provide basic economic elements necessary for states to implement internationally-binding regulations on biodiversity and to promote management actions towards IAS.
Despite some acknowledged methodological flaws (Holmes et al. 2009;Perrings 2011;Cuthbert et al. 2020), Pimentel's studies (Pimentel et al. 2000(Pimentel et al. , 2005 have effectively raised awareness on the grossly underestimated costs associated with alien species introductions (Hensley 2012;Bradshaw et al. 2016). In these works, most of the costs related to invasive species were analysed for the United States. Therefore, invasion costs incurred in Europe have been understudied, despite Europe being a historic centre of globalisation (Reba et al. 2016), and, by virtue, a hub for alien species introductions. Nonetheless, Kettunen et al. (2009) estimated a total cost of approximately 12 billion Euro (€) per year across Europe (Haubrock et al. 2021b). However, many of these cost estimations are untraceable (and thus potentially unreliable) and information at national scales has remained anecdotal at best. Such national-scale information is essential, as it is at this level that budgeting and decision-making are often made.
One particular example is Italy, which has been considered as one of the financial and cultural centres for the development of Europe (James and O'Rourke 2011), with a history of many alien species introductions (Occhipinti-Ambrogi 2002; Nunes et al. 2014Nunes et al. , 2015Tricarico et al. 2018). In total, the Global Register of Introduced and Invasive Species (GRIIS, Pagad et al. 2018; National Database on Alien Species, IS-PRA) lists in excess of 3,000 known alien species in Italy, at present, with 15% of those species considered invasive. These have occurred through various pathways, such as agriculture, angling or horticulture, to name only a few. As such, Italy is today referred to as a hotspot and gateway for several groups of invasive species (Occhipinti-Ambrogi 2002;Occhipinti-Ambrogi et al. 2011;Castaldelli et al. 2013;Nunes et al. 2014Nunes et al. , 2015. In particular, aquatic ecosystems in Italy have been burdened with well-intentioned introduction efforts in support of recreational angling (Occhipinti-Ambrogi 2002; Gherardi et al. 2008;Gravili et al. 2010). This has led to the establishment of more than 150 freshwater aquatic species (at least 64 invertebrates and 48 vertebrates), reported in 2008 to contribute at least 2% of the inland-water fauna (Gherardi et al. 2008;Tricarico E. pers. comm.). In addition, approximately 165 alien marine species have been recorded along the 7,000 km long coastline, aided by increased, human-mediated habitat connectivity (Occhipinti-Ambrogi et al. 2010). In terrestrial ecosystems, at least 923 insect species taxa have been introduced (Inghilesi et al. 2013).
Despite these burgeoning numbers of high-impact invasions in Italy, cost data on the Italian economy are still scarce. The lack of cost quantifications impedes decisionmaking by policy-makers and stakeholders, owing to a distinct absence of an economic rationale for environmental priority actions. We hypothesise that these costs are substantial, although a considerable difference in costs amongst ecosystem types can be expected. Based on data from current literature, we synthesised and described, for the first time, the costs of invasions on the Italian economy. More particularly, we first aimed at depicting how these costs are distributed according to the a) invasive species or broader taxonomic groups (i.e. classes or orders), b) socioeconomic sectors, c) geographic regions and d) cost types, whilst examining the effects of habitat type within each of these descriptors. Second, we determined how the overall costs have changed since cost reporting began and whether these costs of invasions are depicting a particular trend over time.

Methods
To investigate the invasion costs on the Italian economy, we used cost data collected in the InvaCost database (2,419 entries; Diagne et al. 2020a, b) concerning the global costs of invasive species, based on published literature, enabling comprehensive quantification of costs associated with invasive species at various spatial and temporal scales. This updatable database was constructed, based on both published and grey literature and enables the most comprehensive cost quantifications associated with invasive species. All the methodological procedures, from literature searches to data collation, have been detailed elsewhere (Diagne et al. 2020b;Angulo et al. 2021a). All cost entries were standardised to a common and up-to-date currency (US dollars (US$) 2017). We complemented the data following two specific ways: on one hand, we have added cost data collected from non-English documents, including Italian (5,212 entries; Angulo et al. 2021a. https://doi.org/10.6084/m9.figshare.12928136). Further, we added supplementary cost data from new references containing cost information (ca. 2,300 entries; https://doi.org/10.6084/m9.figshare.12928145.v1).
The period of estimation across reported costs varied considerably, spanning periods of several months to several years. For the purpose of the analysis and to derive the total cumulative cost of invasions over time, we considered the duration time (i.e. number of years) over which each cost occurred. For this purpose, we defined the duration of each cost entry. We based this on the difference between the starting ("Probable starting year low margin" column) and ending ("Probable ending year low margin" column) years of the reported costs. When no period of impact was specified in one and/or the other column(s), we counted only a single year unless the authors were certain that the costs had been repeated up to a certain year. The obtained figures corresponded to the total cumulative cost along a defined period for each entry.
From the full database, we identified cost entries related to the Italian economy by filtering data using the 'Official country' column. In addition to the already available information present in the database, we added six further cost records, summarised in the work of one of the authors (Tricarico et al. 2018) and sent them to invacost@updates.fr to be integrated into the InvaCost database. The final dataset can be found in Suppl. material 1. Finally, the invasion costs were specifically estimated from all entries according to method reliability, i.e. indicating the reliability of cost estimates, based on the type of publication and method of estimation. Estimates in peer-reviewed publications or official reports or with documented, repeatable and/or traceable methods were designated as High reliability; all other estimates were designated as Low reliability (Diagne et al. 2020b); the taxonomic group ('Class', 'Order' and 'Species' columns); activity sector (the activity, societal or market sector that was impacted by the cost; 'Impacted sector' column, see Suppl. material 2); invaded habitat ('Environment' column); and lastly the cost type ('Type of cost' column) by grouping costs according to the categories: (a) Damage referring to damages or losses incurred from invasion (e.g. costs for damage repair, resource losses, medical care), (b) Management comprising control-related expenditure (e.g. monitoring, prevention, management, eradication) and money spent on education, research and maintenance costs, (c) Mixed including mixed damage and management costs (cases where reported costs were not clearly distinguished amongst cost types). In addition, for the purpose of investigating the costs in different habitats, we defined costs on wetlands and riparian zones generated by organisms that have an association with both terrestrial and aquatic environments, as Semi-aquatic. For costs that were estimated at a spatial scale below the Country level within Italy, we considered them in a finer-scale Regional analysis, using the 'Spatial scale' and 'Details' columns.
We estimated global average annual costs of invasive species in Italy represented in the InvaCost database by quantifying the temporal trends in cost accumulations. We performed these estimates for the period from 1990 to 2020. To investigate invasion costs in Italy over time and, hence, to identify whether costs are saturating over time or continuously increasing, we used the summarizeCosts function of the 'invacost' R package (Leroy et al. 2021). We thus determined decadal average costs since 1990, as well as the cumulative and average annual cost of that entire time period. Overall, this approach allowed for trends in raw data cost to be examined over time, corresponding to the impacted year when the cost was incurred.

Results
There were 50 economic cost entries (40 of which from the original InvaCost database) associated with Italy. After expansion, the collective 207 expanded database entries totalled US$ 819.76 million between 1990 and 2020 (US$ 26.44 million per year). From these, 76% of costs were actually realised (i.e. assigned to Observed category in the 'Implementation' column) and 97% of the total costs were considered as of High reliability ('Method reliability' column) and, thus, derived from peer-reviewed or traceable sources.

Economic costs by taxonomic group
A high proportion of the filtered database entries concerned invasive mammal species (number of expanded database entries n = 88; US$ 149.81 million, Table 1). Invasive insects represented the second most reported class (n = 57; US$ 273.42 million), followed by invasive Malacostraca (n = 33; US$ 24.45 million). In turn, invasive plants (n = 5) and invasive nematodes (n = 13) had a combined overall cost of US$ 371.72 million. Looking at specific orders, invasive Asterales contributed, with US$ 344.80 million, the most to the cost burden, followed by invasive Coleoptera (US$ 153.81 million), Rodentia (US$ 149.43 million) and Diptera (US$ 116.22 million). Taken together, all other orders accounted for less than US$ 100.00 million (Table 1; Fig. 1).

Economic costs by geographical region
For several publications within the database, specific information about the region where the cost actually occurred was provided. This indicated a difference in economic cost data between the north and the south of Italy. Regional information was present only for northern/central regions, which included Emilia-Romagna (US$ 99.05 million; n = 27), Latium (US$ 24.27 million; n = 27), Trentino (US$ 20.27 million; n = 7), Lombardia (US$ 10.45 million; n = 14), Piedmont (US$ 762.63 thousand; n = 15), Tuscany (US$ 547.21 thousand; n = 17), and Umbria (US$ 22.33 thousand; n = 1). From southern regions, only Sicily had reported costs (US$ 6.70 million; n = 4). Accordingly, there was a distinct lack of cost estimation concerning States in southern Italy. The other entries were given either at national level (US$84.83 million; n = 40) or from unspecified locations (US$572.85 million; n = 55).

Temporal accumulation of costs
The recorded average annual cost between 1990 and 2020 amounted to an average US$ 26.44 million, with an exponential increase in decadal means over time ( Fig. 3; note the y-axis is log 10 -transformed). In the most recent years, recorded costs of invasions in Italy have surpassed US$ 50 million per year.

Discussion
The overall cost of invasive species in Italy has been estimated at US$ 819.76 million (€ 704.78 million) between 1990 and 2020. This can be seen as a conservative total cost  and is most likely an underestimate of the actual costs, especially as costs have been occurring for several decades prior to the first Italian record and because very few known invasive taxa have reported costs. Indeed, many species or invasions have not been studied for their economic impact and many of those that have been studied are not publicly available (e.g. governmental reports, grey literature) and costs present therein were thus not recorded in the InvaCost database. Furthermore, it should be noted that some types of costs are simply difficult to quantify, especially regarding ecosystem services or damages and losses (Spangenberg and Settele 2010). The lack of such data is critical because it can give the false impression that costs due to invasive species are lower than they actually are. Consequently, due to a lack of synthesis of cost information from multiple sources, as well as disparate reporting amongst invasive taxa and invaded regions, decision-makers may not have the necessary information to evaluate the costs and benefits of invasive species management actions. As a result, this limits rationale to invest in preventative or control measures to reduce or manage invasions. However, while we acknowledge that the InvaCost database does not capture all available costs through the systematic searches employed, our data indicate that the economic costs of invasions and studies are growing exponentially over time and are unlikely to saturate soon.
Taxonomic, habitat and regional biases across current literature are prevalent, resulting in only a subset of invasions being evaluated. For example, to our knowledge, costs associated with phytosanitary inspections are not available. Moreover, several aquatic invasive species which are known to damage Italian freshwater ecosystems (e.g. freshwater crayfish) were only anecdotally recorded in the database without identifiable species, given a distinct lack of cost estimation. Additionally, impacts associated with invasive crayfish species, such as levee damaging burrowing behaviour (Haubrock et al. 2019), habitat engineering (Barbaresi et al. 2004), in addition to direct impact on native biodiversity (Gherardi and Acquistapace 2007), were not present in the InvaCost database. This may be due to the cultural and financial benefits of introductions outweighing potential economic cost or the nature of ecological impact, leading to nonvalue costs, which are inherently harder to estimate (e.g. native community changes, ecosystems changes) and, therefore, not reported.
Despite the number of invasive species in Italy exceeding 3000 (Gherardi et al. 2008;Pagad et al. 2018; National Database on Alien Species, ISPRA), the recorded costs of invasive species in Italy concerned only very few species. For example, fish were totally missing, despite known invasions and costs. In addition, reported costs in Italy were lower than in many other EU countries (Haubrock et al. 2021b). Indeed, national-scale costs were an order of magnitude higher in the UK (Cuthbert et al. 2021a), Germany (Haubrock et al. 2021c), France (Renault et al. 2021) and Spain (Angulo et al. 2021b). This could reflect the insufficient investment by the Italian public services into the surveillance and management of these species (i.e. via control, monitoring) relative to the direct impact cost on health and crops (i.e. health care, damage cost), as well as non-market values linked to biodiversity, ecosystem services, well-being or cultural benefits (Plieninger et al. 2013). Our results also show that the highest incurred costs are linked to the control of insects. It does not mean that invasive alien insects are more frequent in Italy, but only that these are seen by public authorities as a priority to limit the damage to agriculture and forestry and the risk of emerging vector-borne human diseases, such as yellow fever, Zika, dengue or chikungunya, which are pathogens vectored principally by Aedes aegypti and Ae. albopictus (Beltrame et al. 2007;Zammarchi et al. 2015). Damage-losses through mammals (i.e. rodents) played only a subordinate role compared to insects. This could be due to several factors, such as lack of dedicated funds and coordinated and continuous control plans, public perception against the control of mammals (especially for squirrels and even for coypu) and problems with legislation (i.e. before EU Regulation 1143/2014, the change of juridical status of the coypu caused a decrease in control actions). Yet, amidst the noted lack of reported costs for insects and mammals, these taxonomic groups may present the complete records amongst all recorded groups.
Nevertheless, the lack of information on invasion-related costs could be due to the search terms used to identify economic cost literature when building InvaCost or to the lack of specific data available. For instance, a search in Italian, using a different search string including species names that are known to cause damage in Italy, yielded two additional cost estimations (as included in our data analysis). This highlights the existence, but restricted availability of important information, as (1) in the Region of Latium, central Italy, economic impacts of alien crayfish species were estimated between €140,000 and 1.17 million per year, including damage to angling, aquaculture and agriculture (Gherardi et al. 2014); (2) in the Region of Piedmont, damage in rice fields due to burrowing activities of Procambarus clarkii led to a decrease of 6% in the annual rice production, while in 'Consorzio di Bonifica dell'Emilia Centrale' €1000 per m 2 was necessary to rebuild levees damaged by its burrowing activity (Gherardi et al. 2014). Similarly, damage caused to agriculture by coypu was estimated at around €1 million, while damage and restoration of levees incurred costs of nearly €11 million (Panzacchi et al. 2007).
Regionally, economic costs of invasive alien species were reportedly higher in the north of Italy (North: US$ 155.37 million; South: US$ 6.70 million). Regions in northern Italy are commonly associated with higher population densities and human activities (Gherardi et al. 2008). Northern Italy also belongs to a different climatic zone (Grapow and Blasi 1998;Celesti-Grapow et al. 2010). It is, however, difficult to be confident that these observed regional cost differences reflect true differences in governance or monetised impacts, rather than a discrepancy in reporting costs between northern and southern regions. That is due to the fact that reporting of costs can be described as limited, considering that only 15 invasive species had recorded costs. Further, a considerable share of the total costs (~ 80%) was not clearly distinguishable for any region, with this low resolution negating a more comprehensive regional analysis. However, extensive/intensive agriculture (CREA 2017) and drainage basins (Gherardi et al. 2008) are concentrated in the north, potentially explaining part of the bias in reported economic costs that were available in InvaCost. As a result, the described lack of aquatic costs entries in the north is a cause for concern, as the numbers of invasive species and affected freshwater ecosystems are relatively high in the north of Italy, indicating a severe neglect of invasive aquatic species costs for Italy overall. The overall bias towards costs on terrestrial habitats and the scarcity of information regarding aquatic habitats could be explained by the nature of the damaged economic sectors with terrestrial ecosystems being more commonly perceived, leading to more public awareness amongst populations. This trend also reflects that on the global scale, whereby aquatic species invasion costs have been underreported compared to terrestrial taxa, relative to known numbers of alien species between those habitat types (Cuthbert et al. 2021b). Additionally, Italy has a long history of active species introduction into freshwater ecosystems. Amongst species like the red swamp crayfish P. clarkii (Gherardi and Acquistapace 2007) or the pond slider Trachemys scripta (Ficetola and Scali 2010), a very well-known example is the black bullhead Ameiurus melas, which first appeared in 1904 (Tortonese 1970) and has now spread throughout Italy, reaching high densities in, for example, the River Tiber and the Corbara Reservoir (Pedicillo et al. 2008).
The key drivers for these introductions have often been cultural, because the introduction of alien freshwater species is often perceived as favourable and beneficial for local municipalities (Selge et al. 2011;Kilian et al. 2012), as it increases the attraction of fisheries, for example, for recreational angling. Therefore, in future studies, it would be important to address the costs of invasions for the economy, whilst considering potential values, particularly in terms of angling.
In addition, we identified an exponential increase in costs through time since 1990, with annual average cost exceeding US$ 50 million in recent years. Given that the US$ 819.76 million total is attributed to only 15 out of the more than 3,000 known alien species in Italy (Pagad et al. 2018; National Database on Alien Species, ISPRA) and probably far from exhaustively for these 15 species (see above), it can be easily appreciated how significantly underestimated this overall cost is, regardless of how they are representative of these few species. For instance, only one database entry referred to the grey squirrel (Sciurus carolinensis), which has been studied and managed thoroughly in Italy (Genovesi and Bertolino 2001;Martinoli et al. 2010) and no entries regarded the well-investigated costs from the impacts of invasive crayfish were found (Gherardi et al. 2014). Further, the costs of management activities in Italy spent solely on IAS within the Union Concern list of EU Regulation 1143/2014, between 2016 and 2018 (in total 48 listed species for this period, 31 of which are present in Italy and 20 that were managed between 2016 and 2018), were estimated at EU€1.85 million (~ US$2.17 million) for those two years (Alonzi et al. 2020). Annually, this figure suggests that the monetary investments into management efforts were between 24-times (considering the entire period) and 46-times (considering only the two recent years) lower than the total costs inferred by IAS at the national level. However, it should be noted that (i) today, 42 of the now 66 IAS listed are present in Italy and, (ii) this cost does not consider the allocated funds for IAS at regional levels. As a result, the actual, unreported or unevaluated economic costs of all IAS in Italy must be staggering.
In conclusion, the presented economic costs of biological invasions in Italy will contribute to informed decision-making at the national level and, thus, providing economic incentives for mitigating the arrival, spread and damage of invasive species. The relatively-high costs reported for Italy, despite the low number of entries in the database contrasting with the high number of invasive species, underlines the need for prevention and surveillance programmes, as costs spent on these are generally considered several magnitudes lower than active management (Leung et al. 2002). Our study highlights the need for national and regional authorities to produce more structured reporting of costs in order to refine these figures further (Diagne et al. 2020b. However, future perspectives indicate an urgently-needed increase in national budgets to tackle the threat of alien species (Silva et al. 2014;OECD 2019). Further studies are also required to examine costs attributed to invasive species in other EU countries, as these may facilitate better-coordinated international actions and drive policy change to mitigate economic costs of invasive species. Nevertheless, we highlighted the increase in annual costs for Italy: the numbers of invasions will increase over time (Seebens et al. 2017) and, thus, it is most likely that reported costs will continue to rapidly increase during the coming decades.