Aim Contemporary dispersal constraints and environmental conditions are broadly recognized as significant drivers of beta diversity patterns. However, beta diversity patterns may also reflect the legacy of past climatic and geological events. In this study, we investigated the relative importance of historical and contemporary factors as drivers of taxonomic, functional and phylogenetic beta diversity in Neotropical stream communities.

Location The Colombian Orinoco basin.

Major taxa studied Diatoms and insects.

Methods We estimated taxonomic, phylogenetic, and functional beta diversity using Baselga’s (BAS) and Podani’s (POD) frameworks. Following both frameworks, we further partitioned every biodiversity facet into turnover and nestedness or richness difference components. Then, we used generalized linear models (GLM) to relate each biodiversity facet with environmental, spatial and historical factors.

Results We found that both historical and contemporary factors affected current patterns of beta diversity. Historical factors and water pH and temperature had the strongest effect on beta diversity patterns, particularly for taxonomic and phylogenetic facets. GLM models performed better for insects than for diatoms in all three facets. Within communities, our analysis also revealed a partial congruence between BAS- and POD-based results.

Main conclusions Due to their past geological history and contemporary environmental gradients, tropical montane streams are natural laboratory for disentangling the joint effects of ecological and biogeographical factors on biodiversity patterns. Our study reveals that present-day distribution patterns cannot be fully explained without accounting for the effects of past geological and climatic events on mountain landscapes. In the Neotropics, montane geology sets the stage for speciation and landscape formation, with which ecological (e.g., dispersal limitation) and environmental factors interact to generate spatial variation in species turnover.

Study area

The Orinoco basin is the third largest basin in South America, covering an area of about 990,000 km2that is in most of Venezuela and in the eastern part of Colombia(Romero Ruz, Galindo Garca, Otero Garca, Armenteras Pascual, 2004). The complex geological and climatic history of the basin has shaped a broad range of ecosystems across heterogeneous landscapes(Romero Ruz et al., 2004).

In total, we sampled 26 (for diatoms) and 32 (for insects) stream segments during the dry season of 2017(January-February). In each stream segment (100 to 200m long), we selected three riffle areas that were representative of the range of substratum types, flow velocities, channel widths and depths, and canopy cover occurring along the stream. Physical and chemical variables were measured during invertebrate sampling (January-February 2017) and on two further occasions (November 2016 and, January-February 2018), that corresponded to high and low water flows, respectively. Instantaneous discharge was estimated in the three riffles by measuring of water depth and flow velocity at 15cm intervals along three cross-sections. At each interval, we also recorded the dominant substrate. Flow velocitywas measured with a digital flow meter (SCHILTKNECHT MiniAir 20). Canopy shading (%) was estimated from vertical photographs using a fisheye lens and subsequent image analysis.Conductivity, pH, oxygen, and temperature were recorded using a HANNA HI98194 water quality meter upon arrival (early morning) and departure (dawn) from the site.

On each occasion, 1 L of water was collected for physico-chemical analyses, filtered through 0.7m glass fiber filters (Whatman GF/F, Kent, UK) and stored frozen until analysis. In the laboratory, ammonium and nitrate concentrations were determined on a Dionex ICS-5000 ion chromatography system (Dionex Corporation, Sunnyvale, U.S.A.). Reactive phosphorus (PRS) concentrations were determined colorimetrically using the fully automated discrete analyzer Smartchem 140 (AMS Allaince, Frpillon, France). Total suspended solids (TSS) were analyzed by filtering 500ml of water through a pre-weighed GFF and drying the filtrate for 1 hour at 105C. The mean and coefficient of variation of all the variables per ecoregion are summarized in Table S1.

Longer-term hydrological variables were estimated using the rational method modified byTmez (2003). This method estimates a streams water flow as function of the total precipitation, the basin area and associated land uses, the time of concentration, and the runoff coefficient (Supplementary Material).Once the daily water flow had been determined, we estimated the threshold at which the streams basal flow was surpassed, as a unit of disturbance for the invertebrate communities.We then calculated: (i) the number of days elapsed since the last flood event (defined as the one doubling the basal flow discharge); (ii) the number of flood events; and (iii) the ratio between the maximum and basal flow discharges.

Invertebrate sampling

Insects were collected using a multi-habitat sampling procedure, with 5 Surber (mesh size: 350mm; area: 0.09 m2) samples collected in stream substrata that were selected according to their corresponding habitat coverage. For instance, if a riffle was composed of 60% of boulders, 30% gravel, and 10% cobbles, 3 Surber samples of the first, 1 of the second, and 1 of the third substratum type were collected.The substratum distribution in each riffle was evaluated visually using the Wentworth scale (mm, diameter-based) as a reference(Wentworth, 1922).We only sampled boulders (diameter 250mm) smaller than the sample frame. In six of the 32 streams sampled, only two riffle sections (10 Surber samples) were sampled because of problems with access.

In the laboratory, invertebrates were sorted and identified to the genus level, followingTrivinho-Strixino Strixino (1995), Merritt Cummins (2008), Domnguez Fernndez (2009) and Gonzlez-Crdoba et al.(2015). Chironomidae and Ephemeroptera were dissected and mounted in Euparal following the protocol ofDomnguez (2006) and Andersen et al. (2013). The pupae of Chironomidae were mounted to confirm some taxonomical identities (Prat et al. 2014).

Diatom sampling

We sampled diatom communities at three different riffle sections of each stream segment, each section spanning from 20 to 60m long.At each riffle section, we collected 8 cm2of surface, brush-scraped algal material from 30 boulders and cobbles. In the case of streams from Guiana shield and high-Plains, where boulders and cobbles were scarce, we also took samples from bedrock, pebbles and sand. Algal material was pooled by riffle section (= 3 samples per stream segment) and subsequently preserved in aTranseausolution. In the laboratory, the organic material from samples was cleaned using hydrogen peroxide. Clean diatom frustules were mounted on permanent slides using a Naphrax medium, the slides were then observed under a 1000x light microscope and identified at the finest level possible using specialized monographs(Krammer and Lange-Bertalot 1986, 1991, Metzeltin and Lange-Bertalot 2007, Bellinger and Sigee 2015). At least 400 valves were counted in each slide.