Functional traits explain both seedling and adult plant spatial patterns in gypsum annual species

Read the free Plain Language Summary for this article on Journal blog. Ecological processes are known to leave recognizable footprints spatial patterns of plants. Observed can be tested against theoretic expectations based effects species dispersal, fine-scale abiotic factors and biotic interactions realized plants (McIntire & Fajardo, 2009). The point-to-point relationships among plants, as ‘plant's eye perspective’ (Murrell et al., 2001), have been successfully used infer heterogeneity interspecific at fine scales perennial (Chacón-Labella 2016; Jara-Guerrero 2015; Pauchard Pescador 2014, 2020). Seed dispersal profoundly affects structure population community levels (Cousens 2008; Wang Smith, 2002). initial pattern provides biological template that allows subsequent ecological act, such germination, herbivory, pathogen infection, facilitation competition, thereby leading distribution seedlings then adults (Nathan Muller-Landau, 2000; Schupp Fuentes, 1995). Depending relative strengths all these processes, will disappear, mainly due non-random thinning 2014; Kenkel, 1988), or retained by adult (e.g. Lara-Romero 1995; Seidler Plotkin, 2006). It is well adapted restrictive soils (i.e. serpentines, gypsum soils) usually behave island edaphisms showing limited seed dispersal. This ensures offspring remain in same similar environmental conditions mother (Friedman Stein, 1980), minimizing risk unsuitable habitats. Fine affect signal plant (Pescador Specifically, soil crusts (BSCs) important sources habitats (Escudero Ortiz 2023). BSCs act filters during bank formation (Peralta 2016), germination seedling establishment 2007; Romão Escudero, 2005) determine annual composition (Luzuriaga 2012; Peralta 2019; Sánchez 2022). also ecosystem functions limiting evapotranspiration increasing availability resources beneath them (Berdugo Delgado-Baquerizo 2013). Consequently, primary source communities soils, they expected deep signals coexisting species. may vary throughout ontogeny depending specific strategies. Functional traits promote different competitiveness stress tolerance capacities (Garnier Westoby 2002), which observed modify functional under 2020) (López-Rubio 2022) systems. In study, we explore, first time our knowledge, relationship between conform local assemblages two climatically contrasting years. allow improve knowledge about role assembly very scales. Key mass, maximum height, leaf area (SLA) reproductive/vegetative biomass ratio (reproductive ratio, hereinafter) clearly connected main assembly. For example, large-seeded competitive (Freckleton Watkinson, 2001; Rees, 1995) advantages McConnaughay Bazzaz, 1987; but see Tracey Aarssen, 2019) compared with small-seeded size positively correlated (Leishman 2000), larger favoured drought Westoby, 1994). height related advantage distance (Thomson 2011), resistance water-limited environments (Dovrat capacity establish hard 2019). Species high SLA reproductive ratios associated more efficient use (Moravcová 2010; 2018), ultimately greater ability (Kraft Pérez-Ramos Coexisting exhibit broad variability their preferences (from opportunistic specialized gypsophytes; Luzuriaga 2015, 2018). fact, it variations trait values ‘limiting similarity hypothesis’; Abrams, 1983, 1996), variation phylogenetic relatedness neighbours (Chaves 2021) a critical mechanism allowing scale coexistence. Annual excellent models evaluating These diverse (around 120 species; 2018) form rich 38 per 0.25 m2 rainy years; 2012, 2015) small-sized individuals (10.5 cm tall average) short synchronized life cycles autumn spring). Thus, every individual mapped relatively small over from stage until individuals. Furthermore, restart cycle year bank, so could evaluate climatic each We measured afore mentioned species, fully patterns, fitted point process elucidate limitation responses Baddeley Lin 2011; Shen analyses were conducted separately stages consecutive years (a dry wet year) experimental set ups: intact BSC versus artificially disturbed plots located central Spain. Here, valuable tool understand an setup where manipulated BSC—known crucial filter communities—during climate conditions. particular, following hypotheses (see Table 1, detailed explanations): (1) (Murren, Reich 2003); (2) Limited occur around would reflected aggregated least stage; (3) should experiment was flat zone top hill El Espartal (Valdemoro, Madrid, Central Spain, 40°11′11.5″N 3°37′47.0″W—permit Ref. 10/058764.9/15 Comunidad de Madrid) (Figure S1). mean precipitation 365 mm year−1 temperature 15°C (www.aemet.es). vegetation dominated shrubs (30% cover) many specialists, Helianthemum squamatum (L.) Dum. Cours., Lepidium subulatum L., Centaurea hyssopifolia Vahl, Gypsophila struthium L. Loefl., Launaea fragilis (Asso) Pau, some generalists, including Thymus zygis Loefl. ex Retama sphaerocarpa Boiss. Macrochloa tenacissima Kunth. perennials scattered well-developed matrix lichens Diploschistes diacapsis (Ach.) Lumbsch, Squamarina lentigera (G.H. Weber) Poelt, Fulgensia subbracteata (Nyl.) Poelt Psora decipiens (Hedw.) Hoffm) 2012). seasonally covered (up 2012) densities 1200 m−2). comprises tiny (mean = 9.7 ± 8.0 cm) complete whole October April. most abundant annuals Asterolinon linum-stellatum Duby DC, Bromus rubens Micropyrum tenellum Link, salicifolium Mill., Ziziphora hispanica strict gypsophytes include Chaenorrhinum reyesii (C. Vicioso Pau) Benedí Campanula fastigiata Dufour DC. Six measuring 1 × m randomly distributed 0.4 ha avoiding holes depressions, microtopographic differences created One plot extended 1.20 ensure 1000 present plot. three plots, kept other thoroughly destroyed mallet remaining fragments powder left surface (‘disturbed BSC’, Figure S1) maintaining total bank. monitored emerged 33% below average 35% above second year, designated ‘dry year’ ‘wet year’, respectively Although both study remarkable, aware non-measured year. sampled twice register (in December) (April). (28,330 individuals) six (two cycles). identified traced rooting DIN-A3 transparent vinyl (polyvinyl chloride, PVC) sheets. mobile methacrylate suspended 10 avoid biocrust damage. total, 216 PVC sheets recorded converted into digital images photo-scanner (Epson Expression 10000XL; Seiko Epson Corp.). Images assembled unique picture Adobe Photoshop CS3 software (Adobe Systems). coordinates plant) ArcGIS 10.1 (ESRI, 2011) using interactive pen display (Wacom DTU-2231). economics spectrum (SLA ratio), reproduction trade-offs (seed mass), hierarchies (maximum height) (Tables S1 S2). estimated healthy undamaged collected spring outside sampling surrounding 2019 details). All according protocols described Cornelissen al. (2003) Pérez-Harguindeguy (2013). 61 years, only considered 242 26 15 dates S3 S4). To characterize pattern, Ripley's K function (Ripley, 1976): λK(r), λ intensity density), number points within circle radius r arbitrary homogeneous pattern. Note cumulative fitting cluster better estimates parameters than non-cumulative pair correlation (Baddeley 2015). four commonly ecology studies (for mathematical details, S2): Poisson (PP) model assumes locations either adults) independent constant across plot, result random environment (Lin 2009); inhomogeneous (IPP) independence recognizes existence assumed microhabitat (Wiegand 2007); (PCP) aggregation (Shen 2009), generated two-step process: PP centres plants) ρ first, before produces distribution, mothers radially symmetric Gaussian zero standard deviation σ, employed measure (Jara-Guerrero 2006; (4) (IPCP) PCP), varies response habitat Moloney, 2014). select best Akaike's information criterion sum residuals Webster McBratney, 1989). K-functions translation edge correction (Ohser, 1983). up 250 steps mm. functions, selection calculated R (R Core Team, packages spatstat 2015), ecespa (de la Cruz, 2008) selectspm generalized linear analyse treatments (intact vs. disturbed), (wet autumn), (seedling adult) frequency types fit models). nonparametric Kruskal–Wallis tests whether determined treatment. PCP (n 156), multiple regression examine traits, treatment, abundance parameter σ (which size). log-transformed normalize data. noted statistical analyses, ignored IPCP because comprised less 1% cases (only performed version 4.0.5 2021). kruskal.test stats package. Multinomial multinom nnet Linear lm package Anova car found 27.5 4.6 2060 1018.4 (1 m). (64%) IPP (31%). nine PPs (3.7%) IPCPs (<1%) (some examples shown treatments, did not significantly system (Table S5). Around 65% maintained 1), 17.2% shifted clustered stage. By contrast, 15% (PP IPP) (PCP IPCP) cycle. trend highly regardless treatment especially significant 2; 2). exhibited predominantly (PCP), whereas lower had distributions IPP). heavier seeds (PP), those smaller tended (PCP). heights taller Finally, PCP, affected 3). Nevertheless, higher sizes (higher values). key determinants minor extent. Our results showed models), agrees lack mechanisms atelechory; sensu Ellner Shmida, 1981), presence mucilaginous coats anchor vicinity myxospermy Engelbrecht García-Fayos, S3). previous analysed standing ephemeral evidence (Olano 2005; Caballero López-Rubio unpublished data). strategy interpreted method likelihood success occupying microenvironments extremely low probability larger-dispersal special (Ellner 1981; García-Fayos Remarkably, dominant support hypothesis edaphic specialists short-distance adaptive appropriate cohort. appeared populations disturbance yearly Indeed, detected stage, contradicts usual findings communities, forests becomes overdispersed roles played density-dependent He Duncan, 1988). maintenance consistent favourable microenvironment near outweighs any possible costs (Ward, 2016). percentage studied (18%) demonstrating mortality occurs norm Mediterranean soils. Spatial explained proportion (32%) (IPCP suggest effect chemical must negligible evidenced microtopography experimentally removed selecting zones meaningful relief BSC, system. However, contrary expectations, mediated physical implies unmeasured properties stages. disaggregated modified fractions remained soil. microscopic portions lichens, algae, fungi bacteria active probably metabolic (Delgado-Baquerizo 2013; Escudero 2007). Similarly, (2023) beneficial growth water development S1c), surprising although effect, its influence BSC. agreement hypothesis, connections traits. mass strongly remark S2); thus, well. Atelochorous lacking mechanisms) dominate S3), determines explain why shorter ~5 formed 8 16 Another complementary explanation predators secondary dispersers heavy light seeds. Small survive longer banks (Peco 2003), enter readily 2005), joining immediately after falling (Baskin Baskin, 1998; Chambers 1991; Thompson, 1987) losses ant predation large trapped particles (Gutterman, 2002; Leishman 1994), thus exposed ants Ant one drivers loss (Azcárate Peco, 2006), has documented preferably 2005). reflect predation. produce trade-off (Reekie hinder well-defined clumps, might simply numerical reasons. A remarkable study. strategies adaptively exploiting resources, 2004; Wright 2004) heterogeneity: highlight Opportunistic cope wide range resource conditions, germinated established microsites primarily dispersed, plant, generating patterns. conservative (low ratios) non-clustered heterogeneous (IPP), underlying microenvironmental retention surface, irrespective integrity rainfall run-off events promoted blur while successful point, however, cannot discard confounding summary, confirmed dynamics, is, ontogeny, suggesting close particularly semiarid Environmental almost one-third contrast Adrián Marcelino Cruz conceived ideas designed methodology; Ana data; Arantzazu led writing manuscript. authors contributed critically drafts gave final approval publication. thank Arpegio Parque Regional del Sureste. Gema Sánchez, Pablo Pino Salvador Romero help measures Angélica Gallego Belén Moncalvillo data digitalization. Spanish Meteorological Agency (AEMET) providing specially Julia Chacón-Labella modelling. A.M.L.P. supported FPU grant (grant no. AP2012-6702). funded projects: AGORA-CGL2016-77417-P Ministerio Economía y Competitividad Government, REMEDINAL TE-CM (S2018/EMT-4338) Madrid GYPWORLD project (H2020-MSCA-RISE-2017-777803—European Commission) PID2020-114851GA-I00 Ciencia e Innovación. Authors no conflicts interest. Data available Dryad public repository link: After publication: https://doi.org/10.5061/dryad.59zw3r2cm. peer review: https://datadryad.org/stash/share/JQ_HlMbgJf3JXtWqYVveWj1Fi3U1i2D_S8vJwGc1G0o. S1: Mean 46 evaluated study: (μg), Specific Leaf Area (SLA, mm2/mg), Reproductive/vegetative (Rep/Veg) Maximum (mm). S2: Correlation coefficients (Spearman's ρ) p-values pairwise correlations: (SLA), height. Bold numbers indicate correlations (p < 0.05). S3: List date. S4: Number date biocrust), adult), (dry autumn). S5: Biocrust frequencies (PP, IPP, PCP). rare (<1%), excluded. Sampling location Spain). Example single marked level (PVC) Please note: publisher responsible content functionality supporting supplied authors. Any queries (other missing content) directed corresponding author article.