Examples of natural or seminatural disturbances: wildfires (a), flooding (b), severe wind storms (c), or drought (d); Source: Wikipedia.org

Contexts in source publication

Context 1

... disrupts ecosystem, community, or population structure and that changes resources, substrate availability or condition, or the physical environment" (White and Pickett 1985, p. 7), have increasingly gained importance among ecologists . This is in part due to the fact that recurrent disturbance events such as floods, fires, or severe windstorms ( Fig. 1) have spurred the evolution of some of the most perplexing life-history strategies in plants, including seed dormancy, resprouting, and vegetative dormancy (Murphy 1968;Benton and Grant 1996). These adaptations translate into life-cycle transitions, i.e., survival, growth, reproduction, and recruitment, cued to disturbance regimes ...

Context 2

... is a short-lived subshrub up to 45 cm tall with circinate, linear leaves grouped in dense rosettes and covered with stalked mucilage-producing glands . It produces large, sulphur-yellow, hermaphrodite flowers, radiate and pentamerous, borne in stalked, cymose inflorescences (Paiva 1997; Correia and Freitas 2002; Figure 1). Flowers are homogamous, i.e., possess a spatial and temporal closeness between dehiscing anthers and receptive stigmas, with high selfing capability even in pre-anthesis ). ...

Context 3

... one of the noticeable features of the highly autogamous Drosophyllum is the production of still large, showy flowers on peduncled inflorescences (Figure 1). Therefore, considering the high allocation costs of flower production (Galen 1999;Andersson 2005), what are the benefits of large, conspicuous flowers in a carnivorous plant species presumably independent of the role of pollinating insects for reproduction (OrtegaOlivencia et al. 1995)? ...

Context 4

... grow in rosettes, and number of rosettes in this species is a good proxy for age. Plants 1-2 rosettes in size initially reproduce in the second year after emergence and gain 1-2 rosettes per previous rosette each growing season ; Figure 1D). ...

Context 5

... et al. in prep.). Bright sulfur-yellow flowers on each scape open gradually and last one day in full anthesis, so that no more than two flowers per rosette are in anthesis at the same time (Figure 1). Flowers are large , with an average petal length of 2.84 ± 0.21 cm and petal width of 1.89 ± 0.17 cm (chapter 2). ...

Context 6

... possible presence of volatile compounds that might attract (or repell) insects in this odourless adhesive was discarded after headspace solid-phase microextraction coupled with gas chromatography- mass spectrometry (HS-SPME-GC-MS). So, plants and mimics looked morphologically and chromatically similar to the human naked eye (Figure 1), but mimics were fully odourless and, although we did not measure reflectance spectra, likely differences in UV reflection patterns may be expected between mimics and plants, since Drosophyllum leaves produce considerable UV reflection ( Joel et al. 1985). Therefore, mimics would function as adequate null models against which insect attraction by Drosophyllum plants may be tested, ...

Context 7

... plants grew more than four times as much as not insect-fed plants during the experiment ( Fig. 1) and produced a more than five times higher dry biomass, both above-and belowground ( Fig. 2), regardless of soil fertility conditions. ...

Context 8

... two-way repeated-measures ANOVA detected significant effects of the two factors, insect feeding and soil fertilization, on relative plant growth (Table 1; Fig. 1). In addition, plant size changed significantly with time (days after sowing), with plants growing significantly faster when fed with flies compared with unfed plants (Table 1; Fig. 1). Correspondingly, the two-way MANOVA showed significant effects of both factors on the final dry biomass of above-(shoot) and belowground (root) ...

Context 9

... two-way repeated-measures ANOVA detected significant effects of the two factors, insect feeding and soil fertilization, on relative plant growth (Table 1; Fig. 1). In addition, plant size changed significantly with time (days after sowing), with plants growing significantly faster when fed with flies compared with unfed plants (Table 1; Fig. 1). Correspondingly, the two-way MANOVA showed significant effects of both factors on the final dry biomass of above-(shoot) and belowground (root) portions of plants, and a significant interaction between the two factors ( Table 2; Fig. 2). The significant interaction effect stemmed from soil fertilization having a slight but ...

Context 10

... and (ii) due to their typically small size, tracking the fates of individual seeds in natural habitats without disrupting the soil is currently a nearly impossible task (Baskin and Baskin 1998; Navarra and Quintana-Ascencio 2012). Consequently, even if data on seed banks are collected, researchers usually extrapolate their long-term fates ( Fig. 1) from short-term field experiments or models (Menges 2000). These approaches are sensitive to parameter uncertainty due to relatively small sample sizes and must account for this uncertainty when estimating population ...

Context 11

... including diapause, vegetative dormancy, or migration) where vital rate quantification from data may contain high uncertainty. Figure 1 Possible fates of seeds after maturation at time t in the studied species Drosophyllum lusitanicum. Mature seeds either germinate and become established as recruits (goCont) the growing season following maturation in t+1 or enter the permanent seed bank (goSB). ...

Context 12

... performed two 3-year field seed burial experiments and a greenhouse germination trial, overall using > 5,100 seeds, to quantify the possible fates of seeds -including seeds in the seed bank, the discrete component in the IPM (Fig. 1). Details on all experiments can be found in Appendix A2. Both field experiments were initiated in September 2012 and 2013, ...

Context 13

... we randomly collected seeds from five populations and buried mesh bags containing 20 seeds each in recently burned and adjacent unburned heathland patches. We then dug out mesh bags 18 months after burial to estimate seed seed-bank stasis (staySB), which consisted of two probabilities: surviving and not germinating from the seed bank ( Fig. 1). We assumed that the proportion of viable seeds encountered after 18 months corresponded to stasis within one time interval in the IPMs (one year), ensuring that seed-bank dynamics were at the same time scale as the rest of the species' life cycle modeled (Appendix S2). In a separate experiment, using the same design as in the mesh-bag ...

Context 14

... < 1 cm below the soil surface. We recorded germination 6 and 18 months after sowing to estimate, respectively, immediate seedling establishment, i.e., the probability of establishment in the spring following seed dispersal (goCont), and the probability of establishment, or egression, from the seed bank at least two springs after dispersal (outSB; Fig. 1). The vital rate outSB consisted of two probabilities that we could not separate: seedling emergence and survival to establishment (Fig. 1). We defined the proportion of seeds entering the seed bank (goSB) as 1-goCont -ω S , where ω S = seedling mortality prior to the census, i.e., seedlings that emerged 4 months after sowing but ...

Context 15

... i.e., the probability of establishment in the spring following seed dispersal (goCont), and the probability of establishment, or egression, from the seed bank at least two springs after dispersal (outSB; Fig. 1). The vital rate outSB consisted of two probabilities that we could not separate: seedling emergence and survival to establishment (Fig. 1). We defined the proportion of seeds entering the seed bank (goSB) as 1-goCont -ω S , where ω S = seedling mortality prior to the census, i.e., seedlings that emerged 4 months after sowing but failed to establish (Appendix S2). Lastly, in greenhouse trials, we exposed seeds to heat and smoke treatments and quantified germination, which ...

Context 16

... with the vital rates describing seed-bank transitions: ingression into (goSB), stasis (staySB), egression from the seed bank (outSB), and both staySB and outSB. We sampled parameters for seed-bank stasis and egression independently because seeds that do not stay in the seed bank may die before successful establishment, i.e., outSB ≠ 1 -staySB (Fig. 1). We kept the remaining vital rate parameters at their average posterior values to assess effects of parameter uncertainty on estimates of population viability of seed-bank related vital rates only (see makeIPM.R in Appendix ...

Context 17

... of population viability are available in sLambdaSimul.R and sLambdaRmpi.R for implementation using parallel processing. Figure 2 Hierarchical structure of simulations of the stochastic population growth rate, log λ s , incorporating parameter uncertainty of three vital rates: seed-bank ingression (goSB), stasis (staySB), and egression (outSB; Fig. 1). Bayesian posterior distributions were sampled to obtain 1,000 parameters for each vital rate. For each parameter, log λ s were simulated from 100 stochastic projections, each run over 4,000 discrete time steps t using Markov chain transitions between five time-since-fire (TSF) environments (0, 1, 2, 3, >3). The transitions depended ...

Context 18

... stasis (staySB) and egression (outSB; Fig. 1) had the largest relative effects on the stochastic growth rate, log λ s , of Drosophyllum populations across fire return intervals (Fig. ...

Context 19

... results showed that life-cycle transitions related to the seed bank ( Fig. 1) chapter 6). Our elasticity analyses suggested that increases in both seed-bank stasis and egression would strongly, positively affect the stochastic population growth rate (Fig. 3). However, these two vital rates are negatively correlated, implying that seed-bank stasis can only be optimized at the expense of egression and vice versa ...

Context 20

... probability of survival, fl[f], is estimated as a Bernoulli distribution, We used normal uninformative priors (μ = 0; 1/θ 2 = 1×10 -6 ) for most fixed factors and for the Gamma-distributed rate parameters, ρ, of the Poisson-Gamma mixture models for the number of flowering stalks (í µí¼‘ 1 ) and number of flowers per stalk (í µí¼‘ 2 ). (Fig. S3.1). The í µí¼ parameters describing the standard deviation in the growth (γ) and seedling-size (φ 4 ) likelihood functions were associated with uniform priors. We used hyperpriors for the random site and block effects ( Fig. S3.1). The hyperpriors were defined as a normal distribution í µí±(0, í µí¼) in which the precision, í µí¼ to ...

Context 21

... mixture models for the number of flowering stalks (í µí¼‘ 1 ) and number of flowers per stalk (í µí¼‘ 2 ). (Fig. S3.1). The í µí¼ parameters describing the standard deviation in the growth (γ) and seedling-size (φ 4 ) likelihood functions were associated with uniform priors. We used hyperpriors for the random site and block effects ( Fig. S3.1). The hyperpriors were defined as a normal distribution í µí±(0, í µí¼) in which the precision, í µí¼ to be estimated using the prior σ ~ í µí±¢í µí±›í µí±–í µí±“(0, 100) for the linear and σ ~ í µí±¢í µí±›í µí±–í µí±“ (0,20) for the logistic regressions, respectively. Details on the priors we used can be found in the R ...

Context 22

... 131 ~ Table S4.1 Differences in proportional contribution of parameter uncertainty to the variation in the stochastic population growth rate of Drosophyllum lusitanicum as a function of vital-rate parameters sampled in stochastic simulations. The simulations were defined by varying fire return intervals, and the vital rates sampled corresponded to ingression, goSB, stasis, staySB, and egression, outSB (Figure 1 in main Figure S4.9 Plots display average log λ s (points) and 2.5 and 97.5 quantiles (vertical lines) as function of fire return interval (x-axis). At each fire return interval, the quantiles were calculated from log λ s estimates obtained from four different approaches: running 100 stochastic projections of log λ s for each of 1,000 posterior parameter samples describing either seed-bank ingression and stasis (black; six parameter); or model parameters describing above-ground survival (blue; 13 parameters); or all parameters related to above-ground dynamics (red; 93 parameters); or running 100 stochastic simulations using mean parameter values for all vital rates (grey ...

Context 23

... Spain, northern Morocco, and western Portugal, differ in the type, frequency, and intensity of anthropogenic disturbances in heathlands ( ; Fig. 1d). ...

Context 24

... 140 ~ Across the species' range, we surveyed 74 sites characterized by nutrient-poor soils and different degrees of disturbances in heathlands ( Fig. 1a (Fig. 1c). We took a subset of 30 out of the 74 locations in southern Spain to fit models for anthropogenic disturbances at the local scale (Fig. 1b). Here, the sampled locations represented (i) a mixture of mature Mediterranean heathlands characterized by dense, low scrub cover ); (ii) marginal habitats in clearances such as ...

Context 25

... 140 ~ Across the species' range, we surveyed 74 sites characterized by nutrient-poor soils and different degrees of disturbances in heathlands ( Fig. 1a (Fig. 1c). We took a subset of 30 out of the 74 locations in southern Spain to fit models for anthropogenic disturbances at the local scale (Fig. 1b). Here, the sampled locations represented (i) a mixture of mature Mediterranean heathlands characterized by dense, low scrub cover ); (ii) marginal habitats in clearances such as firebreaks, road ...

Context 26

... 140 ~ Across the species' range, we surveyed 74 sites characterized by nutrient-poor soils and different degrees of disturbances in heathlands ( Fig. 1a (Fig. 1c). We took a subset of 30 out of the 74 locations in southern Spain to fit models for anthropogenic disturbances at the local scale (Fig. 1b). Here, the sampled locations represented (i) a mixture of mature Mediterranean heathlands characterized by dense, low scrub cover ); (ii) marginal habitats in clearances such as firebreaks, road sides, or on steep slopes characterized by sparse shrub cover; and (iii) managed habitats (e.g., ...

Context 27

... natural heathlands, Drosophyllum has a life cycle typical of a short-lived seeder species (e.g. Menges and Quintana-Ascencio 2004; Fig 1a), germinating from a persistent seed bank after fire . Seed germination is triggered by both heat and removal of surrounding vegetation (Correia and Freitas 2002; Chapter 5). ...

Context 28

... by livestock and game -mimic the removal of shrub canopy by fire (Chapter 6). Populations exposed to human disturbances show a markedly different population structure (longer above-ground survival of adult plants and annual recruitment; Chapter 6) than fire-disturbed heathland populations but experience fire occasionally (Plan INFOCA 2012; Fig. 1). With increasing local extinctions across the species' range (Correia and Freitas 2002), assessing population viability under different combinations of disturbances is not only important for conservation of heathland ~ 166 ~ biodiversity but may also shed light on the potential fate of fire-adapted seeder species in Mediterranean ...

Context 29

... quantified the dynamics of Drosophyllum populations exposed to variable time since last fire and two levels of browsing/trampling pressure by domestic ungulates (Fig 1b). We asked (i) under what conditions disturbance interactions may cause population extinction vs. persistence. ...

Context 30

... 167 ~ Figure 1 The carnivorous subshrub Drosophyllum lusitanicum shows life-cycle adaptations to recurrent fires and occurs in a fire-prone Mediterranean biodiversity hotspot. (a) In the integral projection models, life-cycle transitions (here simplified to stages) were represented by 12 vital rates (blue); Solid and dashed arrows represent transitions of survival [above-ground (σ) or seed bank (staySB, outSB)]/growth [above-ground (γ) or seedling size (φ 4 )] and fecundity [flowering probability (φ 0 ), # stalks (φ 1 ), # flowers/stalk (φ 2 ), # seeds/flower (φ 3 ), immediate germination (goCont), and seed-bank ingression (goSB) or egression (outSB)], respectively; σ S -above-ground seed survival. ...

Context 31

... quantify vital-rate transitions in Drosophyllum, we parameterized integral projection models (IPMs; ) with census and experimental field and laboratory data. We estimated vital rates of individuals with above-ground biomass from five annual censuses (every April 2011-2015) of a total of 2,378 individuals in eight populations (Fig. 1b). The populations differed with respect to (i) livestock pressure (LS) from browsing/trampling, experiencing either high (HLS; human-disturbed populations) or low (LLS; natural fire-disturbed heathlands) pressure; and (ii) time-since-fire (TSF hereafter), between 1 and >26 years ( Fig. 1b; Table S1.1 in Appendix S1). As Mediterranean ...

Context 32

... of a total of 2,378 individuals in eight populations (Fig. 1b). The populations differed with respect to (i) livestock pressure (LS) from browsing/trampling, experiencing either high (HLS; human-disturbed populations) or low (LLS; natural fire-disturbed heathlands) pressure; and (ii) time-since-fire (TSF hereafter), between 1 and >26 years ( Fig. 1b; Table S1.1 in Appendix S1). As Mediterranean heathland habitats do not change significantly in species composition and structure >3 years after fire ), we transformed TSF into a categorical variable consisting of 1, 2, 3, or >3 years since fire. (Fig. 1a). We used plant size = log(# of leaves × length of longest leaf (cm)), after ...

Context 33

... heathlands) pressure; and (ii) time-since-fire (TSF hereafter), between 1 and >26 years ( Fig. 1b; Table S1.1 in Appendix S1). As Mediterranean heathland habitats do not change significantly in species composition and structure >3 years after fire ), we transformed TSF into a categorical variable consisting of 1, 2, 3, or >3 years since fire. (Fig. 1a). We used plant size = log(# of leaves × length of longest leaf (cm)), after model selection for σ, γ, φ 0 and φ 1 , as the continuous state variable in all IPMs (see below). We also quantified above-ground seed survival from the demographic census data as σ S = 1 -flower damage (Appendix S1). We then used this parameter to modify vital ...

Context 34

... performed two field seed-burial experiments and a greenhouse germination trial to quantify seed fates and thereby the discrete, size-independent component of IPMs (Table 1). Overall, > 5,100 seeds were used in the experiments ( Fig. 1a; details in Appendix ...

Context 35

... burned) and vegetated (long unburned) heathlands and estimated seed survival in the soil, i.e., seed-bank stasis (staySB), and seedling establishment, i.e., the probability of establishment in the spring following seed dispersal ~ 169 ~ (goCont) and the probability of egression from the seed bank at least two springs after dispersal (outSB) (Fig. 1a). We defined the proportion of seeds entering the seed bank (goSB) as 1-goCont -ω S , where ω S = seedling mortality prior to the census (Appendix ...

Context 36

... 176 ~ A great deal of uncertainty, largely due to vital-rate parameter estimation (Fig. ...

Context 37

... associated with the mean estimates of log λ s across simulations (Fig 3a, c; Fig. S3.1a, c in Appendix S3). High parameter uncertainty resulted in the 95 % CI of log λ s estimates always including 0 values, implying little certainty in effect of both disturbances on long- term stochastic dynamics (Fig. S3.1a, c in Appendix S3). However, parameter uncertainty did not change the general trend in the response of log λ s and P q (150 years) to varying fire and livestock ...

Context 38

... Computer Corporation, 2013) 1 m above ground within each population to record temperature ( ° C) and relative humidity (%) in hourly intervals from January 2013-December 2015. The data loggers confirmed lack of microclimate differences among sites and years with the exception of summer relative humidity, which was higher at two sites in 2013 ( Fig. S1.2). However, no climatic differences between sites occurred in the critical spring growing season (March-May). ~ 186 ...

Context 39

... were considered the input into the seed bank. Seeds that germinated in the second growing season, 20 months after sowing, were considered germination out of the seed bank. In 2014 and 2015, we recorded seedling emergence in winter (February), in order to estimate the mortality of newly emerged seedlings from emergence to establishment in April (Fig. ...

Context 40

... used the information on flower damage (Fig. S1.3) to estimate the parameter σ S = (1 - flower damage) depicting proportion of seeds surviving above ground. The σ S parameters in HLS habitats were estimated to be σ S = 0.35, 0.35, and 0.37 in TSF 2, 3, and >3, respectively. The equivalent estimates in LLS habitats were σ S = 0.18, 0.20, and 0.18 (Table S.1.3 below). We used our ...

Context 41

... habitats were estimated to be σ S = 0.35, 0.35, and 0.37 in TSF 2, 3, and >3, respectively. The equivalent estimates in LLS habitats were σ S = 0.18, 0.20, and 0.18 (Table S.1.3 below). We used our estimates of seedling mortality before establishment to define seed-bank ingression as 1-goCont -ω S , where ω S = mortality in the seedling stage ( Fig. S1.4). We averaged ω S across all treatments so that this rate corresponded to 0.03 in all ...

Context 42

... 188 ~ Figure S1.4 Mean (± S.E.) mortality of 50 sowed seeds as seedlings (before establishment as recruits) as function of burned and unburned habitat patches and two years in which mortality was measured. ...

Context 43

... seed-bank analyses conducted in the spring of 2012, in which we collected 20 random soil samples from the eight study populations and counted viable Drosophyllum seeds in the samples, indicated depleted seed banks in early post-fire habitats (Fig. ...

Context 44

... our estimates of seed dynamics (Table S1.3), we also simulated the number of seedlings produced at each TSF category assuming different initial numbers of seeds in the seed bank. Our simulations agreed with the range of seedling numbers observed during demographic censuses (Fig. S1.6). ~ 191 ~ Table S1.2 Summary statistics of demographic data based on five years of censuses of the carnivorous plant Drosophyllum lusitanicum at eight sites. The mean (± S.E.) vital rates of continuous stages (above-ground biomass) are provided for habitats experiencing high (HLS; white background) and low (LLS; grey background) ...

Context 45

... quantify the extent of facilitation and/or competition by shrubs during post-fire habitat succession, we carried out two experiments replicated at two Mediterranean heathland sites within the Aljibe Mountains, at the northern side of the Strait of Gibraltar (Fig. 1). Parts of the two study sites burned by wildfires (see below). Natural Drosophyllum populations occur at both sites, but were located > 200 m away from the experimental settings. used in this study and experimental design at each study site. At Sierra Carbonera, the seed-sowing experiment was designed as random plots (P) in a burned ...

Context 46

... test the interspecific effects of resprouting shrubs on the vital rates germination, survival, growth, and reproduction of Drosophyllum individuals, we conducted a seed- sowing experiment at a burned heathland site (Sierra Carbonera; fire in August 2011). We established seven paired plots perpendicular to the main elevation gradient (Fig. 1) and sowed two cohorts of seeds, in August 2012 and in 2013, to track the aforementioned vital rates. The experiment did not assess germination in response to direct fire cues but rather relative germination in response to indirect cues (chapter 6) in distinct microhabitats during early stages of habitat succession. We therefore did not ...

Context 47

... each of the seven plots, we distinguished two types of microhabitats: open and shrub ( Fig. 1). We chose the most abundant shrub species in each plot as shrub microhabitat, which were either Erica scoparia or Stauracanthus boivini. Both species had similar, rounded/conical crowns and show similar growth rates after fire (M. Paniw, unpubl.), and we assumed that neighbor identity would not significantly affect our results In three ...

Context 48

... with decalcified water) at the University of Cadiz from seeds collected randomly in five natural populations from southern Spain (Fig. 1). We initially performed this experiment in early May 2013, within the growing season of natural Drosophyllum populations, at a heathland site in Sierra Retin ( Drosophyllum individuals were watered daily with 100 mL of decalcified water during seven days to prevent desiccation. We then took the individuals to the laboratory for ...