Coiba National Park is in the Gulf of Chiriquí (7°29'0.3372"N, 81°47'45.2976"W). The Park contains an area of 2701.25 km² (Fig. 1) and the main island is the largest island in the Central American Pacific, with an area of 503.14 km². In addition, the Park contains 38 island territories, such as Jicarón (20.02 km²), Jicarita (1.25 km²), Canales de Afuera (2.40 km²), Afuerita (0.27 km²), Pájaros (0.45 km²), Uva (2.57 km²), Brincanco (3.30 km²), Coibita (2.42 km²), along with several smaller islands, islets and rocky promontories (Autoridad Nacional del Ambiente et al. 2009). It is characterised by moderately high temperatures, averaging 25.9ºC. Its climate is humid tropical with precipitation averaging around 3334 mm³ and an average relative humidity of over 80% (Poorter 1993, Autoridad Nacional del Ambiente et al. 2009).

Figure 1.  

Sampling sites in Coiba National Park depicting the different vegetation types. The square shows the location of Coiba National Park in Panama (left) and a close-up view (right).

This study was carried out between August 2021 and August 2023, when a total of 15 field expeditions were conducted, visiting 13 locations in Coiba and nearby islands (Fig. 1, Table 1). We did not make distinction between 'mature mixed broadleaf forest' and the 'secondary mixed broadleaf forest' (Table 1). The logistical difficulties to travel and stay in the different islands made it difficult to develop a more systematic protocol to monitor elements of population or community dynamics; therefore, our effort focused on creating a basic checklist of species. For the collection of spiders, ad hoc active searching was employed involving day (~ 7:00 - 11:00 AM) and night (~ 7:00 PM - 11:00 PM) walks along pre-existing trails and roads. The most visited sites were the locations 2, 3, 4, 5, 8, 9 and 10 where we carried out five visits at each. We performed three visits to sites 11 and 13, two visits to site 6 and 12 and one visit to sites 1 and 7. We inspected trees, cavities, rocks, human-built structures and other elements to maximise collection opportunities. We used entomological beating trays, sweep nets, plus direct tube collection or aspirators to sample spiders at various heights. In addition, we performed a few canopy searches by using equipment for arborism to climb the trees (~ 15 m). We also used pitfall traps to capture wandering and understorey spiders. Ten traps were placed only in three trails in Coiba (sites 8, 9 and 10), approximately 10 m apart. The traps consisted of transparent plastic containers (diameter: 11 cm, height: 8 cm), containing a 400 ml solution of 70% ethanol and soapy water in a 2:1 ratio. We recovered the traps after 5 days per site visited and samples were transfered to 70% ethanol. Overall, collections from all methods were merged as combined records.

The samples were preserved in 70% ethanol and are stored in the laboratory of Dr Dumas Gálvez at the Central American Master Program in Entomology at the University of Panama. All samples are available for future study on request. We carried out identifications by means of published guides in scientific articles, books, websites and, in a few instances, by direct communication with specialists of different groups (Suppl. material 1). Spider families were classified according to foraging patterns in their natural habitat (Cardoso et al. 2011, Rose 2022). The definition of a guild is based on the ecological characteristics of the species (or higher taxa) that determine the exchange of resources. As largely generalist predators of arthropods, the most important resource for spiders is their arthropod prey and their most important ecological characteristics are probably their feeding method, the variety of prey they hunt, vertical stratification, circadian activity, size body and phenology (Uetz et al. 1999, Cardoso et al. 2011). Finally, we built the rarefaction curve in R (R Core Team 2024) with the function iNEXT and ggiNEX in the package iNEXT (Hsieh et al. 2016).