A comprehensive assessment of the intertidal biodiversity along the Portuguese coast in the early 2000s

Abstract Background The unprecedented rates of current biodiversity loss have motivated a renewed interest in environmental and biodiversity monitoring. The need for sustained monitoring strategies has prompted not only the establisment of new long-term monitoring programmes, but also the rescue of data from historical or otherwise archived sources. Amongst the most valuable datasets are those containing information on intertidal systems, as they are particularly well suited for studying the biological effects of climate change. The Portuguese rocky coast is quite interesting for studying the effects of climate change on the distribution of species due to its geographical orientation, latitudinal patterns in temperature, species richness, species' distribution patterns and availability of historical information. This work aims at providing a comprehensive picture of the distribution and abundance of intertidal macro-invertebrates and macro-algae along the Portuguese rocky coast in the early 2000s. New information This study provides a description of the rocky shore intertidal biodiversity of the mainland Portuguese coast in the early 2000s. The spatial distribution and semi-quantitative abundance of a total of 238 taxa were assessed at 49 wave-exposed locations. These data provide a comprehensive baseline against which biodiversity changes can be effectively and objectively evaluated.


Introduction
Biodiversity is now declining globally at rates unprecedented in human history (Ceballos et al. 2017, IPBES 2019. This process is being mainly driven by habitat degradation and loss and compounded by the recent climate change (IPCC 2014). Biodiversity loss has been severely affecting species distributions, community structure, ecosystem function, ecosystem services, food security and public health (Edwards and Richardson 2004, Hawkins et al. 2008, Wernberg et al. 2011, IPBES 2019. This so-called "biodiversity crisis" has motivated a renewed interest in effective environmental and biodiversity monitoring (Navarro et al. 2017, CBD 2010. Broad-scale, sustained observations, including historical data rescue, are now regarded as essential to understand past trends and to produce accurate forecasts which are needed to provide information for policy decisions (Hawkins et al. 2013).
Intertidal systems are well suited for studies focusing on the effects of climate variability and climate change on biodiversity, as they are amongst the most thermally complex environments on Earth (Pincebourde et al. 2008) and are strongly influenced by meteorological conditions (Bates et al. 2018, Zamir et al. 2018. They are inhabited by marine organisms that must withstand terrestrial conditions during low tide. Stressful events may have dire consequences for these species (Jurgens et al. 2015), which are, therefore, regarded as sensitive indicators of climate variability and change (Helmuth et al. 2006).
Stemming from its peculiar geographic, climatic and oceanographic setting, the biogeography of the Portuguese rocky coast is quite interesting and offers exceptional conditions for studying the effects of climate change on the distribution of species. First, the coast is mostly linear with a north-to-south orientation and with a latitudinal gradient in temperature during the winter, from relatively cold water in the north to relatively warm water in the south. In the summer, that latitudinal gradient is often intensified by the effects of coastal upwelling, which brings deep cold water to the surface near the coast, especially in the northern portion of the country (Lemos and Pires 2004, Lima et al. 2007, Seabra et al. 2015. These clinal variations in temperature result in a region of contact between Monte Clérigo, Arrifana, Castelejo, Martinhal, Ingrina, Praia da Luz, Porto de Mós, Dona Ana, Castelo and Olhos de Água. The location of Buarcos, roughly at the mid-point between the northern and the central stretches of rocky coast, was also included in this study. Sampling description: Each of the 49 studied locations were extensively surveyed from the splash fringe level in the high intertidal (area of occurrence of littorinids and lichens) to the low fringe level at the low intertidal (area of occurrence of red, green and brown algae, see Fig. 2). At each location, a two-people team worked from one hour before the low tide peak to one hour after low tide (Fig. 3). The average astronomical low tide height during  Table 1.

Figure 2.
Examples of shores and tide levels surveyed in the present study. A -Mindelo, in northern Portugal, B -Santa Cruz, in central Portugal and, C -Dona Ana, in southern Portugal. D -High intertidal at Monte Clérigo, with Melarhaphe neritoides, E -Mid-intertidal at Arrifana, with barnacles and mussels, F -Low intertidal at Mindelo featuring Himanthalia elongata, which is now almost extinct from the area. Photos taken by Fernando P. Lima.
surveys was -1.51 0.14 m below mean sea water level (Table 1). The occurrence and abundance of all easily-identified taxa (animals and algae, sensu latu) were recorded in situ. A semi-quantitative estimation of abundance was assigned to each taxa identified during the survey. We used a modified version of the scale established by Crisp and Southward (1958) -SACFOR, where abundances were encoded from 6 to 0 (where 6 means Superabundant; 5, Abundant; 4, Common; 3, Frequent; 2, Occasional; 1, Rare; and 0, not found). Small animals or algae (turfs) or other taxa of dubious classification were collected and their identification finalised in the lab under a stereomicroscope (Fig. 3). Additionally, whole substrate samples were collected by scraping the substrate with a paint scraper spatula at three tidal levels: (1) amongst barnacles, (2) amongst mussels and honeycomb worm reefs (Sabellaria alveolata) and (3) amongst red, green and brown algae and preserved in a solution of 4% formaldehyde in seawater (Fig. 3). The scraped area was approximately 150-225 cm per sample. The number of samples varied between four and six, accordingly to the spatial heterogeneity of each substrate, but totalling approximately 900 cm per shore and substrate. Later, in the lab, formaldehyde was removed from samples by washing them with running water and smaller organisms were separated from larger mussels, honeycomb worm reefs or canopy algae using a 0.25 cm mesh sieve and identified to the lowest taxonomic level possible.
Step description: The steps that led to the final release of the dataset were as follows: (1) In-situ identification of species and attribution of a semi-quantitative abundance SACFOR score; (2) destructive sampling (substrate scraping) at three tidal levels and preservation of samples in formaldehyde; (3) cleaning of formaldehyde, sorting and identification of specimens in the lab; (4) conversion of paper-based records from the field and from the lab into an electronic data format (speadsheets); (5) integration of the field and laboratory datasets into a standardised format; (6) retrieval of missing geographical information, georeferencing of coordinates through Google Earth and general quality control; (7) standardisation of taxonomy against the World Register of Marine Species; (8) export of data as a DarwinCore Archive and (9) generation of dataset-level metadata.

Geographic coverage
Description: Sampling was done along the three major rocky stretches of the entire coast of mainland Portugal, covering an extension of approximately 700 km from Moledo do Minho to Olhos de Água.
Coordinates: 37.021090 and 41.841824 Latitude; -9.486673 and -8.188497 Longitude. The data presented in this paper derives from visual and destructive surveys done along the Portuguese coast in the early 2000s. The dataset published in GBIF has the structure of a Sampling event dataset with two data subsets: Events (Core) and Associated occurrences. These data have been published as a Darwin Core Archive (DwCA), which is a standardised format for sharing biodiversity data. The Sampling Event (Core) contains 49 records (eventID). The extension data (Associated Occurrences) sheet has 11662 occurrences.

Additional information
A total of 238 taxa (Table 2) were identified: 99 Plantae, 36 Chromista, 100 Animalia, two Fungi and one Bacterium (Pereira et al. 2021). A description of number of taxa of each Phylum per location is present in Table 3. The site with the least amount of species was Adraga and the one with the most was Nazaré, both sites in central Portugal. On average, the surveyed locations in northern Portugal had a higher number of species (67 species per location), followed by the locations in southern Portugal (66 species per site). On average, locations in central Portugal (59 species per site) had the lowest number of species (Table 3).   Table 3.