Massive yet grossly underestimated global costs of invasive insects

AbstractInsects have presented human society with some of its greatest development challenges by spreading diseases, consuming crops and damaging infrastructure. Despite the massive human and financial toll of invasive insects, cost estimates of their impacts remain sporadic, spatially incomplete and of questionable quality. Here we compile a comprehensive database of economic costs of invasive insects. Taking all reported goods and service estimates, invasive insects cost a minimum of US$70.0 billion per year globally, while associated health costs exceed US$6.9 billion per year. Total costs rise as the number of estimate increases, although many of the worst costs have already been estimated (especially those related to human health). A lack of dedicated studies, especially for reproducible goods and service estimates, implies gross underestimation of global costs. Globally, insect pests have been reported to reduce agricultural yields by 10 16% before harvest, and to consume a similar amount following harvest9. In fact, the largest food producing countries, China and the United States, exhibit the highest potential losses from invasive insects10. Several other insect pests defoliate trees4 and degrade plant biodiversity, threaten commercial forestry and hamper climate change mitigation via increased tree mortality and associated increases in greenhouse gas emissions11. Many other insects are nuisance species or disease vectors that directly erode public health from the Seventeenth to Twentieth centuries, insect borne diseases caused more human disease and death than all other causes combined12.Insects are also among the most pervasive of invasive species. For example, 87% of the 2,500 non native terrestrial invertebrates in Europe are insects13. Yet, reliable estimates of their impacts are difficult to obtain, in particular for economic assessments. Most cost estimates are disparate, regionally focused, cover variable periods and are not always grounded in verifiable data (see Methods). The types of costs also vary and include both direct and indirect components (Fig. 1). Consequently, extrapolating local costs to global scales is challenging and few have attempted to overcome the many inherent flaws in this approach. Owing mainly to a lack of monetary estimates, we could not compile costs for the categories and subcategories coloured in grey. The inner circle (darkest colours) encapsulates costs associated with prevention; the middle circle (mid range colours) includes costs associated with damage from invasive insects; the outer circle (lightest colours) covers costs associated with responses or follow up to invasive insect incursions. The outermost purple arrow indicates the general increase in our ability to estimate monetizable costs, and the direct relevance to human commerce and well being. DALY, disability adjusted life year (lifespan lost because of burden of insect borne disease; not assessed).Reliable global cost summaries therefore remain a major challenge. Indeed, there are currently only 86 insect species listed in the International Union for Conservation of Nature (IUCN) Global Invasive Species Database14, and of those there are no cost estimates for 81.4%, while 12.8% of them have insufficient (unsourced) estimates. We therefore compiled the most comprehensive database of economic costs for invasive insects available to date (737 screened articles, chapters and reports), standardizing historical estimates as annual 2014 US dollars (US$; Methods).We determined the reproducibility of each study's cost estimates by identifying the source of all values used to extrapolate regional costs. When values were based on actual measures as opposed to non sourced estimates and had a clear methodology provided, we deemed the resulting costs 'reproducible' (although we did not assess quality per se because of a lack of standard, objective criteria to assess the accuracy of published estimates; Methods). We categorized studies that did not meet these criteria as 'irreproducible'. We further divided all costs into two main categories: 'goods and services' (including production of agricultural and forestry goods, and cultural services; Fig. 1) and 'human health', further splitting the former into agriculture, forestry, infrastructure, mixed or urban categories, and the latter into seven disease categories (Methods).Taking all reported goods and services estimates, and avoiding the extrapolation of limited data, invasive insects cost a minimum of US$70.0 billion per year globally, while associated health costs exceed US$6.9 billion per year. Total costs rise as the number of estimates increases; therefore, the true costs of invasive insects to human society are substantially larger (but by a currently unquantifiable amount) than we report here. Further, future costs are likely to increase as invasive insects expand their ranges in response to climate change, as well as to increasing human movements and international trade.ResultsGoods and servicesWe determined that invasive insects cost a minimum of US$70.0 billion per year globally for goods and services, of which US$25.2 billion per year comes from reproducible studies (Fig. 2 and Supplementary Data 1). There was replica van cleef earrings no temporal pattern in annual cost rates (Supplementary Fig. 1), and most estimates were direct measures (although estimated costs were higher for extrapolated costs; see 'Expenditure types and targets' in the Supplementary Methods and Supplementary Fig. 2). Regionally, North America reported the highest annual costs (>US$27.3 billion), followed by Europe (US$3.6 billion per year; Fig. 2a,b), although this is likely more a function of the intensity of research effort (see 'Research effort' below) rather than a true reflection of relative regional costs. The 10 costliest species change little whether including all or only reproducible estimates (Fig. 2e,f). The first column includes all estimates regardless of reproducibility (a,c,e), whereas the second only includes costs for which estimates can be verified ('reproducible'; b,d,f). All costs expressed as annual 2014 US dollars. Bracketed numbers in the x axis labels indicate the number of estimates per category.According to a single study5, the most expensive insect is purportedly the Formosan subterranean termite Coptotermes formosanus estimated at >US$30.2 billion per year globally (Fig. 2e). However, that irreproducible estimate is based on a single non sourced value of US$2.2 billion per year for the United States of America, a personal communication supporting a ratio of 1:4 of control:repair costs in a single US city (New Orleans) and an unvalidated assumption that the US costs represent 50% of the global total5. A more realistic ranking based on the reproducible estimates only (Fig. 2f) places the diamondback moth Plutella xylostella as the most expensive (US$4.6 billion per year)15. Other costly insects include the brown spruce longhorn beetle Tetropium fuscum (US$4.5 billion per year in Canada), the gypsy moth Lymantria dispar (US$3.2 billion per year in North America) and the Asian long horned beetle Anoplophora glabripennis (US$3.0 billion per year in North America and Europe; Fig. 2f).Human healthGlobal health costs directly attributable to invasive insects exceed US$6.9 billion per year (Fig. 3); however, these exclude malaria costs because that disease is not due to the invasion of an insect vector throughout most of its distribution (although malaria cases 'imported' into non endemic areas do incur treatment and prophylaxis costs16). Our summary also excludes the economic impacts on productivity, income, tourism, blood supply system, personal protection and quality of life (Supplementary Note 1), as well as historical epidemics of yellow fever and dengue because no relevant cost estimates exist (Methods). Most health related estimates are a combination of direct and indirect costs (79% and 93% for all estimates and reproducible only estimates, respectively; see Supplementary Note 1), represent actual estimates as opposed to extrapolations or model predictions (66% and 77%, respectively) and are primarily related to medical care (75% and 88%, respectively; see 'Expenditure types and targets' in the Supplementary Methods and Supplementary Figs 3 and 4). Dengue (from a virus transmitted by Aedes albopictus and Ae. aegypti) costs represent 84% of total health costs, followed by 15% for West Nile virus transmitted by Culex spp. (Fig. 3c,d). Asia (US$2.84 billion) and North America (US$2.06 billion) and Central/South America (US$1.85 billion) recorded the highest annual health costs (Fig. 3a,b). The first column includes all estimates regardless of reproducibility (a,c), whereas the second only includes costs for which estimates can be verified ('reproducible'; b,d). All costs expressed as annual 2014 US dollars. Bracketed numbers in the x axis labels indicate the number of estimates per category.Research effortThe regional summaries for both goods and services and health costs belie a strong positive relationship between total costs and the number of individual estimates (see 'Sampling bias' in the Supplementary Methods and Supplementary Fig. 5). Across regions, goods and services costs increase by 10 times for each additional 5.5 (reproducible only) or 13.0 (all) estimates (Supplementary Fig. 5a,b). This strong positive relationship remains when expressed across species (Supplementary Fig. 6), but is necessarily more variable, given that most species have only one cost estimate each. The same type of relationship also exists for health costs, with total costs increasing by 10 times for each additional 18.5 19.1 estimates (Supplementary Fig. 6c,d). This regional bias in sampling corroborates the established phenomenon of a spatial mismatch between invader impacts on threatened species and research publications17, suggesting that large additional costs because of invasive insects remain to be estimated in lesser sampled regions of the world, and reinforcing our hypothesis that the total costs have been grossly underestimated.Cumulative costsGiven that the regions to which these sums apply do not have the same spatial area, have different climates, have important crop and infrastructure differences, and are likely to experience different insect invasion and detection probabilities, extrapolating regional costs to correct for potential undersampling is dubious. We therefore expressed total costs and the number of associated estimates as temporally cumulative values to identify possible thresholds within the sampled regions and categories (see 'Sampling bias' in the Supplementary Methods and Fig. 4). For both global goods and services and human health costs, there was evidence for an asymptote among the sampled species based on fitted logistic models (Fig. 4); however, reproducible only goods and services costs had more support for a non asymptotic linear model (Fig. 4b). This asymptotic behaviour is driven principally by North American goods and services costs (Supplementary Table 1 and Supplementary Fig. 7); in contrast, asymptotic behaviour was more prevalent across compared regions for human health costs (Supplementary Table 1 and Supplementary Fig. 8). For human health costs dominated by those associated with dengue fever, potential undersampling appears less problematic than for clearly underestimated costs from reproducible studies of goods and services. This variable asymptotic behaviour means that only some regions and cost types demonstrate possible evidence of decelerating accumulation rates (that is, the costliest insects are assessed initially, with smaller damages estimated thereafter). For a given year t, we summed all values (costs and number of estimates) up to t (see 'Sampling bias' in the Supplementary Methods for model fitting and comparison methods). We fitted linear, exponential, logarithmic and logistic models to each curve to examine evidence for asymptotic behaviour (identified by the dominance of a logarithmic or logistic model). For all categories except reproducible only goods and services costs (b), the logistic model (curvilinear grey dashed lines) had the highest Akaike's information criterion (AIC) weights (wAICrelative model probability) and explained >96% of the deviance in the data (%DEcoefficient of determination). For reproducible only goods and services costs (b), the linear model (straight grey dashed line) had the highest wAIC, indicating that the logistic asymptote was likely an underestimate. For each fit, we also show the approximate asymptotic cost and the associated number of cumulative estimates required to achieve the asymptote (red lines). See also Supplementary Figs 7 and 8 for accumulation curves expressed by region. All costs expressed as 2014 US dollars.DiscussionThe estimated total global costs, even after attempting to correct for sampling bias, are therefore necessarily gross underestimates. We found only 86 (goods and services) and 117 (health) estimates globally, of which only 55% of the former (n=47) and 85% of the latter (n=99) we deemed reproducible. Ecosystem regulating services, which have high economic value worldwide18, are notoriously difficult to estimate19; hence, estimating the cost of their erosion arising from invasive insects is still unknown. In fact, we identified only one study20 that provided reproducible economic costs of the erosion of ecosystem regulating services (that is, costs not directly associated with goods and services or health, such as the erosion of pollination; Fig. 1) because of invasive insects (two Vespula wasps in New Zealand). That study showed that damages arising mainly from reduced pollination are comparable to the direct costs to goods and services (for example, lost apicultural production and control) and are much higher than associated health costs20.While many non native species are clearly beneficial to human society (Supplementary Fig. 9) by providing food, fibre, ecosystem services and even ecological benefits (habitats and resources for native species21; ex situ conservation22; increasing reproductive success of native plants23), the net outcome from non native insects is strongly negative. This net outcome arises because most invasive non native insects are not directly consumed or used in any way by humans, and their overall benefits to society remain limited24.There are two main phenomena leading to an increased frequency of introductions and potentially expanding distributions of the costliest insect invaders: international trade25 and global warming9. Invasions and subsequent expansions are exacerbated by rising human populations, movement, migration, wealth and international trade25, despite more national and international policies targeting invasive species26. Climate change projections to 2050 also predict a net average increase of 18% in the area of occurrence of current arthropod invaders27.Given that available economic estimates are sporadic, spatially incomplete (especially outside Europe and North America), of variable reproducibility and are likely to increase as the planet warms and international trade expands, we conclude that the costs of invasive insects to human society are underestimated and will escalate with time. The available data describe only the costliest insects of mainly industrial and/or biosecurity concern, and non market costs are rarely estimated (but see Supplementary Note 1), even though they can at times exceed market costs (for example, for forest pests28). In contrast, summaries of direct costs at the scale of the broader economy might not always adequately capture the true net costs of invasive insects because some investments can potentially lead to savings arising from mitigation (for example, costs of purchasing pesticides resulting in reduced damage from targeted pests). It is therefore difficult to estimate total costs from different values of direct and indirect categories of invasive insects impacts; therefore, we recommend that cost summaries always be reported by type and target (for example, Supplementary Figs 2 and 3).Effective, early response and vigilant biosecurity are often cheaper (by up to 10 times for mosquito borne disease29) than waiting to pay for accrued damages4,9, although this might not always be the case when prevention investment occurs long before any impacts are experienced30. In the rare cases where those responsible for novel invasions are identified, 'polluter pays' legislation has been proposed31. However, most costs appear to be borne ultimately by individuals via out of pocket expenses32, higher consumer prices and taxes to fund management31, thus reinforcing the poverty illness nexus33. In addition to improving guidelines for estimating the full costs of invasive insects, vigilant planning, public awareness campaigns and community participation could potentially relieve society of billions of dollars of annual expense, and reduce the contribution of invasive insects to human suffering.MethodsLiterature reviewWe van cleef clover earrings replica began our review of the literature on the economic impacts of invasive insects using the ISI Web of Science database with a specific search string to identify relevant papers (see below). We then used the Web of Science's 'refine' function to restrict the studies identified to the relevant fields, yielding 488 sources from 1911 to January 2014. We analysed each source to reject irrelevant papers and retained those containing economic estimates. We completed our database with 267 relevant studies up to December 2015 (including grey literature) opportunistically gathered. In total, we screened 737 sources, 470 of which were relevant to the economic impacts of invasive insects and from which 158 yielded useable economic estimates (Supplementary Data 1 and 2). When economic values were cited from studies not already included in the database, we searched and gathered papers, reports or chapters providing the initial estimates. For each value, we extracted the estimation methodology, and spatial and temporal coverage (full databases available in Supplementary Data 1 and 2). Owing to the diversity of the methods reviewed, we classified the reproducibility of economic values as 'reproducible' or 'irreproducible' based on qualitative criteria because of the diversity of methods reviewed (see 'Determining cost estimate reproducibility' below). We attributed 'reproducible' to values with demonstrated calculation methodologies, including uncertainties, and with available original references. 'Irreproducible' values were those without calculation methodologies, uncertainty estimates or unavailable original references (see 'Determining cost estimate reproducibility' below). We expressed all costs in 2014 US$34. We averaged multiple values (for example, to provide an annual average over a specified period) or uncertainty ranges before conversion to 2014 US$. In many cases we deemed some of the multiple estimates for the same invasive insect species/disease and region as redundant (that is, generally older, obsolete, incomplete or irreproducible estimates). If monetary costs were provided as a range, we used the median value for each estimate. Detailed calculations for each estimate are available in Supplementary Data 1 and 2.For health costs, we limited the criteria for invasive vector borne diseases and their related vector mosquitoes following Juliano and Lounibos35, based on several van cleef earrings imitation life history traits such as desiccation resistant eggs, development in small, human made containers, occupying human dominated habitats, diapause and autogeny. We added chikungunya and zika and excluded historical epidemics of yellow fever, dengue and malaria in South America because no estimates of these exist. Most of the economic estimates of invasive mosquito borne diseases that we obtained concerned dengue, while only a few concerned Wes

Previous post     
     Next post
     Blog home

The Wall

No comments
You need to sign in to comment