**2. Methodology**

**1. Introduction**

268 Selected Studies in Biodiversity

cancer [10].

preserving insect diversity.

The highest extinction risk and consequently the greatest biodiversity loss are predicted to occur in invertebrates [1, 2], specifically insects, and these declines are expected to cascade onto ecosystem functioning and human well-being [2]. For this reason, mapping ecosystem services provided by insects can be a way to raise people's awareness on the importance of

From 1 million species described, only 5000 are considered harmful to crops, livestock, or humans [3], the so-called dis-services [4]. However, it also widely acknowledge the benefits that insects might directly or indirectly bring to humans, namely through pollination, pest control, raw materials, or as food, known as ecosystem services (ES). Moreover, there are clear evidences of the growing importance of insects as feed and food resources, especially in the context of the global population growth and to overcome the global scarcity of proteins [5] or even in the context of new drugs discovery for by pharmaceutical industries, e.g., [6, 7, 8], with insects having a prominent role in the provisioning services by ecosystems. Insects produce a huge number of defensive secretions against predators, and these molecules are potentially medicinal drugs, e.g., [7, 9]. In countries such as China, about 77 species from 14 families and belonging to 8 orders have been used in traditional medicine to treat tumors and

Due to the growing interest regarding this class as key mediators of ES, i.e., there is a growing demand for ES mediated by insects; there is also the need to assess and map the supply of ES that is mediated by insects and provided in the habitats they populate. The demand for ecosystem services is socio-economically driven [11], while the supply is underpinned by ecosystems process and functions [12]. In this context, this work proposes to deliver a set of potential global indicators for ES mediated by insects that can serve as a foundation for mapping and assessment of ES at local/regional, national, or even transnational levels. The use of indicators has been recognized as a useful communication tool that facilitates the simplification of human-environmental systems high complexity, e.g., [13]; they can be selected for mapping and assessment of ES [14] and to support specific management purposes involving ES trade-offs [15]. This work may also be seen as a contribution to the EU Biodiversity Strategy for 2020, Action 5—Improve knowledge of ecosystems and their services in the EU, which is linked to other global initiatives to protect biodiversity, namely International Platform of Biodiversity and Ecosystem Services (IPBES) and to the UN Natural Capital Accounting. In this sense, we performed an extensive review of the available literature to assess the supply side of ecosystem services, which results directly or indirectly from insect-mediated processes and functions. The ES classification followed the Common International Classification of Ecosystem Services (CICES) that considers three major categories: provisioning, regulating and maintenance, and cultural services in order to avoid natural capital double counting [16]. CICES has been applied to habitats, e.g., [17], to social-ecological systems, e.g., [18], or with the purpose of supporting the mapping and assessment of ES at EU Member States (Action 5 of the Biodiversity Strategy), e.g., [14]. Results from this rule-based review are presented and

discussed according to the lowest possible CICES category.

The ES provided by insects was categorized according to the CICES classification [16], which the latest full version can be downloaded at http://cices.eu/. In this review, we applied the CICES rationale to the world's most abundant and diverse group of organisms and acknowledge their role as major agents of the ES provided in the ecosystems where they live. Following the same principle as the MAES working group [14], we provide a set of potential indicators for each identified service, which might help underpin the mapping, valuation, and management of the delivered services.

The services were categorized to the maximum possible resolution following CICES hierarchical structure whenever possible. If sufficient data were not available, the next higher level of resolution was chosen.

We used a rule-based approach to map indicators for ES. We did this by first reviewing and summarizing existing scientific and nonscientific literature related to ES provided by insects. Our aim was to extract from the literature and by using CICES rational, to classify potential indicators that can be used to assess and map these services in different ecosystems and across different geographic regions.

The literature search was performed in specific scientific libraries (e.g., Scopus and Web of Knowledge), using the keywords (single and in combination): ES, insects, ecological function, pollination, seed predation, biological control, decomposition, seed dispersal, food, feed, materials, medicines, society, and ecosystems. However, there were many gaps for several ES categories in these referenced libraries. For this reason, we enlarged our search to include books and nonresearch literature. This approach was especially important to access the cultural services, for which available scientific literature is scarce. In fact, evidences seem to show that inclusion of nonresearch literature can be useful to validate the results of a research-based literature search [19]. In our case, we definitely would have biased our conclusions if we had excluded nonresearch-based literature, even if this only makes up ~25% of all searched literature, mainly because cultural services are rarely considered in ES assessments [20]. Actually, cultural services do not represent purely ecological phenomena but rather are the outcome of complex and dynamic relationships between ecosystems and humans [21], and for this reason, we needed to adapt our bibliographic review in order to include cultural services and produce valid potential indicators. In total, we used 90 papers indexed in Scopus, 21 authored and 1 edited books, 16 chapters, 13 reports, 7 proceedings, 8 papers not indexed in Scopus, 7 electronic articles, 3 thesis, and 8 webpages.
