Diatom diversity and distribution in Madeira Island streams (Portugal)

Abstract Background Here, we present the data obtained from the samples collected in a field campaign during the spring of 2015 which aims for a better understanding of the diversity and distribution patterns of freshwater diatoms in Madeira Island. Following European and Portuguese standards and recommendations for routine diatom sampling and analysis, we collected samples in 40 sites, distributed in 27 permanent streams and identified the diatom species present, using general diatom floras and studies in Portuguese freshwater diatoms. New information Little is known about the diversity and distribution of freshwater diatom assemblages from Madeira Archipelago. This study reports a survey in 40 sites in Madeira Island distributed in 27 permanent streams. A total of 965 diatom (Bacillariophyta) occurrences were recorded, belonging to 130 different taxa from 44 genera and 27 families. The families with the highest number of occurrences were Bacillariaceae (176), Achnanthidiaceae (135) and Naviculaceae (133). The two diatom endemisms, described previously in Madeira Island (Lange-Bertalot 1993), Nitzschia macaronesica Lange-Bertalot and Navicula madeirensis Lange-Bertalot, were only observed in a small number of sites, located mostly at Laurissilva forest. Sixty species are new records, not only to Madeira Island, but also to the Madeira Archipelago.


Introduction
Diatoms (Bacillariophyta) are microscopic algae and one of the most abundant and diverse group of aquatic, pigmented single-celled photosynthetic eukaryotes, which can be found in almost every type of aquatic environment around the globe (Kociolek 2007, Pla et al. 2016. These microalgae are characterised by an outer silica wall (frustule) that makes them easy to collect and preserve for later identification. Benthic diatoms, in particular, are important contributors to primary production in streams and are widely used as indicators for monitoring the ecological status of aquatic systems and also for past environmental and climatic reconstructions (Battarbee et al. 2006), since many species have distinct ecological optima and narrow tolerance (Cohen 2003). Diatom communities inhabiting streams have been studied in several regions of the globe (e.g. Tison et al. 2005, Grenier et al. 2006, Mora et al. 2017, Falasco et al. 2016, Liu et al. 2019, including insular streams (Delgado et al. 2012, Delgado et al. 2013, Kopalová et al. 2011, Kopalová and van de Vijver 2013, Gonçalves et al. 2015, Vasselon et al. 2017. Despite their great importance, current knowledge about the freshwater diatoms on insular streams of Madeira Archipelago is limited in comparison with freshwater macroinvertebrates (Hughes et al. 1998, Hughes and Furse 2001, Hughes 2006, bryophytes (Sérgio and Fontinha 1994, Sérgio et al. 2006, Sim-Sim et al. 2008, Sim-Sim et al. 2014, Sim-Sim et al. 2011, Boch et al. 2019) and hyphomycetes communities (Raposeiro et al. 2020). Although diatoms from the Madeira Archipelago have been a matter of study for more than 150 years (Grunow 1867, De-Toni 1891, De-Toni 1892, Zimmermann 1909, Zimmermann 1911, Schodduyn 1927, Mölder 1947, Lange-Bertalot 1993, Kaufmann et al. 2015, Gonçalves et al. 2016, including the description of two regional endemisms, little is known about the regional overall diversity of these microalgae in Madeira Island. The importance of insular freshwater studies of microalgal diversity is centred around the concept that these ecosystems tend to be less complex, providing much potential for testing ideas about biogeographic theory and species distribution limits (Flower 2005).
Here, we provide a detailed dataset that contains freshwater diatom occurrences collected during a field campaign on Madeira Island, increasing the knowledge on the epilithic diatom inhabiting permanent streams in Madeira Island. Our purpose is to release this valuable dataset, since no similar datasets have been previously published for Madeira Archipelago and it constitutes a relevant tool of comparison for aquatic ecologists, for example, biogeographic patterns, climate change or other studies on oceanic islands.

Project description
Title: Diatom diversity and distribution in Madeira Island streams (Portugal) Personnel: Collections were undertaken and occurrence data recorded during the spring of 2015 in Madeira Island. The collectors were Pedro Raposeiro and Vitor Gonçalves. Identifications were made by Catarina Ritter and supervised by Vitor Gonçalves.

Study area description:
The Madeira Archipelago is an oceanic archipelago located in the North Atlantic between latitudes 32°24' and 33°07'N and longitudes 16°16' and 17°16'W ( Fig. 1). Madeira Island is the highest (Pico Ruivo -1861 m) and largest island (~ 740 km ) of the archipelago and about 90% of its area is higher than 500 m above sea level (Ribeiro 1985). Madeira Island presents a high diversity of habitat types, including the largest surviving area of Laurissilva forest in Macaronesia, classified as a UNESCO World Natural Heritage site (IUCN 1999). Due to its oceanic condition, Madeira Island presents a mild temperate oceanic climate strongly influenced by winds from the NE and the Canary Islands current, presenting a relative humidity between 55-75% and annual rainfall between 500 and 1,000 mm (AEMET & IM 2012). An important aspect of the climate in Madeira Island is the persistent nebulous covering of fog, which normally exists in high altitude resulting in an important source of groundwater recharge (Prada et al. 2005). Under this mild temperate oceanic climate, groundwater hydrology is essential for surface water and for the persistence and functioning of the insular aquatic ecosystems as a high number of the permanent streams are fed by springs. Madeira Island comprises approximately 126 catchments and 200 streams presenting a typical radial drainage pattern common in oceanic islands (Marques 1994). According to Prada et al. 2005, the hydrographic network present in the Island is characterised by deep narrow valleys with a typical U-transverse profile as these are still in a young phase. Most of the streams have a torrential character with high flow rates (Hughes 2006 (INAG 2008). Epilithic diatoms were taken from stones with a toothbrush in each sampling site (Fig. 5). Immediately after collection, diatom samples were fixed with formalin at 4% final concentration. Permanent slides were prepared with Naphrax® and at least 400 valves per sample were counted and identified at the lowest taxonomic level possible under oilimmersion phase contrast light microscopy using a Leica DM2500 (Leica Microsystems GmbH, Welzlar, Germany).     Nomenclatural and taxonomic status used here follows Algaebase (Guiry and Guiry 2020).
Step description: The data have been published as a Darwin Core Archive (DwC-A), which is a standardised format for sharing biodiversity data as a set of one or more data tables. The core data specificEpithet The name of the first or species epithet of the scientificName. infraspecificEpithet The name of the lowest or terminal infraspecific epithet of the scientificName, excluding any rank designation. taxonRank The taxonomic rank of the most specific name in the scientificName.
coordinateUncertaintyInMetres The indicator for the accuracy of the coordinate location in metres, described as the radius of a circle around the stated point location.

Orders Families Genera Total taxa Total species New records Madeira endemisms
Aulacoseirales  The subphylum Coscinodiscophytina, represented by one class, two orders and three families, accounted for 1.5% of the total occurrences, while the subphylum Bacillariophytina registered 98.4% of the total occurrences. With two classes and four subclasses, most occurrences (815) were registered in the Bacillariophycidae subclass.

Achnanthidium minutissimum (Kützing) Czarnecki and
In this survey, 60 records were new, not only to Madeira Island, but also to the Madeira Archipelago (Table 3). These include 55 species, two varieties and three genera (sp.).

Discussion
The diatom diversity (130 taxa belonging to 44 genera) displayed by Madeira Island in the 27 permanent streams in this study is due to the habitat complexity (including water quality, habitat structure and climate), as well as large scale-effects stemming from the Islands' isolation and geographical location as was found on other oceanic islands (Flower 2005, Gonçalves et al. 2015. Diatom diversity from Madeira Island is relatively low when compared to other oceanic islands and continental regions (e.g. Herlory et al. 2013, Gonçalves et al. 2015, Jyrkänkallio-Mikkola et al. 2018). This kind of comparison is difficult to make since it depends on the sampling efforts: the number of samples analysed, the timing of the samplings, the number of surveys carried out, the physical and chemical composition of the waters, the number of substrates sampled and the taxonomic effort with which the diatom valves were analysed (Morales et al. 2009, Veselá andJohansen 2009).
Nonetheless, when comparing archipelagos in the Macaronesia Region, 201 diatom taxa were recorded in 316 samples from 14 permanent Azorean streams (Gonçalves et al. 2015). However, the number of diatom taxa recorded per sample was higher in Madeira Island (24.1 ± 1.1 SE) than in the Azores (20.9 ± 2.7 SE). Additionally, it is worth mentioning that this study is focused on the main island from the Madeira Archipelago which does not represent the different microhabitats present in all the other islands.
Comparisons to other regions in the world reveal how diatom diversity in Madeira Island is "poor". For instance, in a tropic region (Sub-Saharan Africa), the number of diatom taxa identified in 67 sites in Kenya was significantly greater (297 taxa) than the number of taxa recorded in Madeira Island (Jyrkänkallio-Mikkola et al. 2018).
The low diversity of freshwater biota has already been reported to the Madeira Archipelago (Hughes and Malmqvist 2005), as well as other oceanic islands in the world (Brasher et al. 2004, Flower 2005, Delgado et al. 2012.
This insular oceanic ecosystem should offer some degree of isolation from continental floras, but the special conditions that promote speciation on islands are not present, for example, extreme water quality or geological age and activity (Flower 2005). It is therefore unsurprising that the great majority of taxa had a cosmopolitan distribution (e.g. Achnanthidium minutissimum, Planothidium lanceolatum, Nitzschia soratensis, Cocconeis placentula var. euglypta) (Flower 2005, Gonçalves et al. 2015. This indicates a lack of isolation mechanisms operating in the Island, but the diatom taxa present are nevertheless clearly attributable to different biogeographical regions. The Macaronesia Region should register more endemisms for many groups as do other regions (e.g. Antarctic Region, , Van De Vijver et al. 2004, Kopalová et al. 2009, Kopalová et al. 2011, but due to the lack of research in this Region as stated before, only a few have been found. This might be one of the reasons why we had some taxonomic challenges. Diatoms are extremely diverse and there are many species that have not been described yet, thus the species delimitation is still controversial (Licea et al. 2016). For some diatom groups, it was difficult to distinguish closely-related species because of their wide morphological variation or because many taxa frequently differed in detail from published descriptions. We took a conservative attitude for these taxonomic differences and disregarded small differences in morphometric data.
Considering the significant differences between the islands of Madeira Archipelago, such as geological ages, volcanic composition, climate patterns and distribution, land uses, types of forest and orography, we expect higher diatom species richness and exclusive taxa from these islands (Porto Santo, Desertas, Selvagens). Furthermore, increasing the sampling effort in Madeira Island, for instance, by sampling other streams and/or other sites in the Laurissilva forest, may result in the identification of other diatom taxa. Additionally, higher time replication and larger datasets are required to better understand distribution patterns and large-scale spatial patterns of species dispersal. According to Smucker and Vis (2011), there is a significant under-estimation of diatom diversity when exclusively collecting epilithic habitats for documenting species distribution and for conservation purposes. Taking this into consideration, different methods to obtain the greater number of diatom taxa for each site are encouraged, for instance, by following two distinct sampling techniques (transects located in riffles and from microhabitats) (Falasco et al. 2016).
The factors controlling taxa richness, as well as the regional endemic taxa, remain unclear as there is no convincing link with the simplified habitat features recorded during the study and more research is needed. Relationships are probably multivariate in nature and include site history, unmeasured micro-habitat availability and climate (Flower 2005, Pajunen et al. 2016.
The results of this study provide baseline knowledge on the current distribution of freshwater diatoms on Madeira Island streams, revealing a distinct, but taxonomically simple diatom flora, typical from oceanic island ecosystems (Flower 2005, Gonçalves et al. 2015. In order to better understand the complexity of these streams, depth studies on the temporal and spatial distribution patterns, population dynamics, species' interactions, guilds and traits are essential for improving knowledge and the development of future effective monitoring and conservation programmes and measures for local stakeholders.