Identification of organisms is traditionally based on the use of “classic” identification keys, normally printed on paper. These keys have several drawbacks: they are mainly based on the systematics, requiring identification of orders, families and genera at first; they are written by experts for other experts, in a specific scientific jargon; they have a “frozen” structure (sequence of theses/antitheses); once published, they cannot be changed or updated without printing a new edition. Due to the use of computers, it is now possible to build new digital identification tools, which: 1) can be produced automatically, if the characters are stored in a database; 2) can be freed from the traditional systematics, giving priority to easy-to-observe characters, incl. those usually uncommon to the classical keys, such as ecology and distribution; 3) can be updated in real time once published on-line; 4) can be available on different media, and on mobile devices. An important feature of these new digital tools is their “collaborative” nature. They can be enriched by the contribution of several researchers, which can cooperate while maintaining rights and property of the resources and data they contribute to the system. JellyWeb, the information system on Scyphozoa, Cubozoa and Staurozoa has been developed in Trieste since 2010. The system was created with the aim of – potentially – becoming a starting point for a wide collaborative effort in developing a user-friendly worldwide digital identification system for jellyfishes.

Biodiversity informatics, Cnidaria , FRIDA, identification, jellyfish, Medusozoa

Since the Rio Earth Summit in 1992, access to biodiversity information has become a fundamental task. Biodiversity data are targeted by several efforts of digitalization and aggregation, most of which focus on primary biodiversity data, i.e. natural history collection specimens and field records. Some of these efforts produced wide global networks, e.g. the GBIF (Global Biodiversity Information Facility; Berendsohn et al. 2010, King et al. 2010), which, together with the BioCASE (Biodiversity Collection Access for Europe, Holetschek et al. 2012), is mobilizing ca. 600 millions of records. Primary biodiversity data are mostly used in modeling the distribution of the taxa, and in predicting the effect of climate changes and anthropic pressure on endangered or alien invasive taxa. Taxon related information (nomenclature, auto-ecology, etc.) become the focus of similar large scale efforts only in the last years (Martellos and Attorre 2012, Martellos 2014). The GBIF itself is starting to aggregate checklist data (GBIF 2010), while other efforts are focused on molecular data (Field et al. 2011, Holetschek et al. 2012, Wieczorek et al. 2012), and to ecological information (Fegraus et al. 2005). In the field of hydrobiology, some recent examples can be Fish-SPRICH (Brosse et al. 2013) and Fish-AMAZBOL (Carvajal-Vallejos et al. 2014). In the case of jellyfishes, online resources are however scarce, but some relevant exceptions (e.g., the Jellyfish Dataset Initiative, http://www.bco-dmo.org/dataset/526852).

Digital identification keys are a particular case in the world of biodiversity informatics. Since the development of the DELTA language (Dallwitz 1980), efforts aiming at creating online digital identification keys followed several approaches. The resulting products differ in usability, accessibility, size, etc. (Nimis and Martellos 2009, Hagedorn et al. 2010, Randlane et al. 2010, Martellos and Nimis 2015). With the development of FRIDA (FRiendly IDentificAtion, Martellos 2010), the researchers of the Department of Life Sciences, University of Trieste, aimed at producing a simple but effective instrument for the development of digital identification keys in collaborative efforts. This led to the publication – in the framework of project Dryades, and of the EU projects KeyToNature (http://www.keytonature.eu), SiiT (http://ww.siit.eu) and CSMON-LIFE (LIFE13 ENV/IT/842, http://www.csmon-life.eu) – of ca. 600 different digital identification keys for several groups of organisms.

As far as Scyphozoa, Cubozoa and Staurozoa are concerned, there are digital databases hosting taxonomic information, such as WoRMS (World Register of Marine Species, http://www.marinespecies.org/), as well as paper printed keys to genera (as an example, see Cornelius 1997; other keys are listed in Morandini et al., 2005). Few examples of digital resources are available in the web, often limited to specific geographic regions, as the Cubozoan and scyphozoan key of the Carolinian Biogeographic Province (Calder 2009), the key to the Scyphozoa and Cubozoa of the South Atlantic Bight (Calder and King 2008), and, as far as the Mediterranean is concerned, the web site MeteoMeduse (Boero 2013, http://meteomeduse.focus.it/). The latter, however, is an example of citizen science observatory, and does not provide an identification key. To our knowledge, no comprehensive digital identification tools to species of these taxa exist.

By combining taxonomical, ecological, and morphological and anatomical features into an information system, we developed the so called JellyWeb, a simple tool which allow to researchers and laypersons to identify Scyphozoa, Cubozoa and Staurozoa to the species level. This paper presents the results of this effort, available online at the URL http://dryades.units.it/jelly.

Data were collected from several sources in literature. The most relevant are Kramp (1961), WoRMS (http://www.marinespecies.org/), the Scyphozoan Wiki (http://thescyphozoan.ucmerced.edu/), and Mills (1999-). Further sources are listed in Balboni 2008, Benci 2008, Sarto 2009, Sola 2009, Coral 2012, Benci 2012, Savonitto 2012, Ukosich 2014. Other paper are under consideration, and will lead to adding to the database other species for several genera, such as Atolla (A. russelli, A. gigantea, A. chuni), Aurelia (A. marginalis), Chironex (C. yamaguti), Cyanea (C. lamarkii, C. rosea, C. annaskala, C. tzetlinii, and several other species), Desmonema (D. comatum, D. scoresbyanna), Drymonema (D. gorgo, D. larsoni), Nausithoe (N. marginata), Pelagia (P. benovici) Tripedalia (T. binata).

The information system is freely available online at the URL http://dryades.units.it/jelly. It organizes data collected in the last five years by the research unit headed by Massimo Avian, at the Dept. of Life Sciences of the University of Trieste. The researchers which contributed to the project agreed on distributing the data under a Creative Commons, share alike, by attribution 3.0 (CC 3.0 by-sa) license.

The software of the information system has been developed in PHP language. The data are stored in a MySQL database. The system is equipped with a multi-entry query interface (Hagedorn et al. 2010), which operates on both a taxonomic database, and a database of nine easily recognizable morphological characters (see below). The multi-entry interface allows complex queries, which can be a first step in the identification of an organism. The multi-entry query system returns lists of taxa, on which the identification process can continue by using a digital identification system. The latter has been developed by using the FRIDA (FRiendly IDentificAtion) package (Martellos 2010). It operates on a morpho-anatomical database, which hosts ca. 200 characters for several infra-generic taxa of Scyphozoa, Cubozoa and Staurozoa (a revision of the content of the database due to recent taxonomic advancements is ongoing). The output of the digital identification system is a digital identification key to the remaining taxa, which can be used by an interactive interface, or printed out as a dichotomous, illustrated key. The whole key can also be exported in a stand-alone version for mobile devices (Nimis et al. 2012).

The query interfaces have been developed according to the results of several usability tests, conducted in the framework of projects KeyToNature and SiiT, as detailed in Martellos and Nimis (2015). The system is under continuous development, following users’ input.

JellyWeb hosts several information pages and two query system. The home page (http://dryades.units.it/jelly) provides access to several sections: information, describing how the system works; survey area; query (detailed below); checklist, listing all taxa alphabetically by genus and species name, and providing access to their taxon pages; credits.

The query system is made of two parts.

1. Multi-entry interface (Fig. 1). The first interface of the query system provides the users with the opportunity of specifying a set of nine easily observable characters, and/or scientific name and family. The morphological characters are:

   • Jellyfish sessile / swimming;

   • Umbrella shaped like a cube or a box / not shaped like a cube or a box;

   • Tentacles present / absent;

   • Tentacles isolated / grouped in clusters;

   • Umbrella with a coronal groove / without a coronal groove;

   • Umbrella flat / not flat;

   • Oral arms absent / 4 / more than 4;

   • Jellyfish with filaments (oral arm appendages) / without filaments;

   • Jellyfish with scapulae / without scapulae.

   For each character, an information popup window with images and text detailing the most relevat features is accessible by clicking on the question mark button. The result of a query is a list of taxa (Fig. 2). For each taxon an image is displayed (if available, see below). A link provides access to the taxon page (Fig. 3), which displays a description, as well as all the images available in the system, with credits and metadata, and other information (when available). Taxon pages can host a virtually unlimited amount of information and images, and/or provide access to external resources through HTML links.

2. Digital identification key. The results page of the multi-entry interface allows to generate an interactive identification key to remaining taxa. The key can be used through a simple single entry interface (Fig. 4, Hagedorn et al. 2010), or printed out as a textual, illustrated dichotomous key. At each step of the identification process users can list out the remaining taxa, or print an illustrated key. By clicking on a taxon name, the corresponding taxon page is shown (Fig. 3). Each key generated by this system is different from the others, since they contain a different number of infra-generic taxa. Normally, the lower the number of taxa is, the easier the resulting key. A key to all the taxa currently included in our databases can also be generated, and is provided below.

Figure 1. 

Multi-entry interface. The multi-entry interface allows to combine the states of several morphologic and anatomic characters, together with taxonomic information, to query the database.

Figure 2. 

List of taxa. The result of a query made by using the multi-entry interface is an illustrated list of infra-generic taxa.

Figure 3. 

Taxon page. A typical taxon page displays an image, a description, as well as all the other images available in the system, together with credits and metadata. Taxon pages can host a virtually unlimited amount of data, links and media.

Figure 4. 

Single entry digital identification key. The digital identification key to remaining taxa is generated from the results of the multy-entry query system. It is used through a single entry interface, and can be printed out as a textual, illustrated dichotomous key as well.

Digital resources on biodiversity can be relevant not only to researchers, but also to laypeople, such as tourists or citizen scientists. The importance of involving citizens in understanding, monitoring and protecting biodiversity has been recently expressed by the European Commission, in the document “Establishing Horizon 2020” (EU Regulation no. 1291/2013). However, most of the biodiversity-related resources available in the Web – especially the ones dedicated to to “difficult” groups, such as jellyfish – are normally devoted almost exclusively to experts (Martellos and Nimis 2015). Exposing scientific information in a form which can be accessible to everybody – without losing its content and informative value – can be a true revolution. Many citizens, especially if already interested in nature and aware of environmental issues (e.g. the presence of invasive alien species), are potentially interested in similar resources. Hence digital resources can be used to involve a wider amount of citizens in scientific tasks, such as the collection of those “big data” which are nowadays fundamental to researchers. The examples of OPAL initiative in the British Isles (http://www.opalexplorenature.org; accessed 08 August 2015) or, in the field of jellyfish, of MeteoMedusa (Boero 2013, Boero et al. 2013), and JellyWatch (http://www.jellywatch.org/; accessed 08 August 2015) are demonstrating the effectiveness of a citizen science approach in collecting scientific data.

JellyWeb is based on morpho-anatomic and taxonomic data, collected and organized in ca. 10 years of research. The development of the portal (Martellos and Nimis 2015) was based upon the experience of the European project KeyToNature (mainly devoted to digital identification) and of the project Dryades (devoted to the pubblication of biodiversity data in the web). This is the first portal devoted to organisms other than vascular plants developed by the research unit of the Dept. of Life Science of the University of Trieste. During its development, a particular attention was paid to user interfaces, in order to provide high quality scientific information in the most straightforward way, and to make it useable by the wider audience as possible.

The multi-entry interface can be useful to both researchers (whom can simply type the name of a taxon to retrieve related information or generate an identification key), and laypeople (whom can use it to start the identification of a jellyfish they have just seen on the seashore). As a further help, interactive keys are enriched by images and drawings of the most relevant characters. Since digital keys are generated in real time, on the basis of the list of remaining organisms, each query produces a different identification key.

Since identification is nowadays often based on molecular analysis, the system has been developed to host molecular data as well. In fact, several attempts to revise the taxonomy of the various taxa like the Discomedusae on the basis of morphological observations integrated with genetic analysis are underway, highlighting several critical points, such as the recognition of cryptic species in the Aurelia complex within the “traditional” species Aurelia aurita (Dawson and Jacobs 2001, Dawson and Martin 2001, Dawson 2003, Dawson et al. 2005, Ramšak et al. 2012), or even at higher taxonomic levels like the proposition of at least two new families within the Semaeostomeae(Bayha and Dawson 2010, Straehler-Pohl et al. 2011). The integration of molecular information in a digital identification system by using the FRIDA software was studied by Bruni et al. (2012) for vascular plants.

JellyWeb is an accumulative system, which can potentially host all data on Scyphozoa, Cubozoa and Staurozoa, and even extend its aim to other groups of the phylum Cnidaria. However, a research group alone can hardly complete such a challenging task. The research unit at the University of Trieste plans to maintain and enrich JellyWeb, but its growth could be faster, if other research groups join this effort. A researcher, or a research group, can contribute to the system by:

• Fostering a taxon (such as a genus, or a family). This can be done by managing an instance of the FRIDA system. FRIDA allows to different authors to independently manage separate instances, while at the same time contributing to the same database of morphological ans anatomical data, hence, generating updated multi-authored keys to any subset of taxa in the whole system (for a complete description see Martellos 2010). All the digital keys which are generated by the system give credit to the authors of all the data. The keys and all the data and images in JellyWeb are always distributed under a Creative Commons share alike, by attribution 3.0 license (CC 3.0 by-sa).

• Contributing to the image archive. High quality images of morphological and anatomical characters and of the whole organisms are probably the most relevant bottlenecks in the process of creating a portal such as JellyWeb. Especially when identifing a taxon, digital images are of capital relevance, both for choosing among the leads of each choice, and as visual census when an identification has been achieved. Several species of Scyphozoa, Cubozoa and Staurozoa are known for one or few specimens, and, even when the taxa are well known, high quality images are, however, scarce. JellyWeb was developed to host a virtually unlimited number of images for each taxon. Each image is displayed with credits to the author(s) and owner(s), institution(s), other metadata, and license.

• Producing descriptions. Another relevant bottleneck in developing digital identification keys and portals to one or more groups of organisms are their descriptions. While taxonomic descriptions can be found in books and papers, descriptions which could be actually useful to people other than researches are difficult to produce. In our experience, to be appreciated by a wider audience, they should mix different sources of information, from ecology to taxonomy, from distribution to human uses, relevance for economy, etymology of the name, etc. Hence, their production is not a simple cut and paste, but a relevant effort of analysis and synthesis.

Potential contributor can contact Massimo Avian (avian@units.it), to define the extent of their participation.

The authors are grateful to all the researchers and students (Balboni, Benci, Coral, Sarto, Savonitto, Sola) who contributed to this study. We are also grateful to Dr. Rodolfo Riccamboni, who developed the code of the multi-entry interface of the portal, and to Prof. Pierluigi Nimis, for his suggestions and comments. We are also grateful do Mrs. Sara Triulzi for English language check.