Richness, diversity, and rate of primary succession over 20 year in tropical coastal dunes
The tropical coastal dunes in central Gulf of Mexico have been stabilizing over the last decades resulting in reduced substrate mobility, and promoting primary succession. We describe changes in species richness and diversity in dune vegetation during 20 years. Our questions: (a) Do species richness and diversity increase over time as predicted by models of ecological succession or do they show a hump-backed manner similar to the observations in temperate coastal dunes?, (b) What is the interaction between vegetation cover and diversity and species richness?, (c) Is there a relationship between species diversity and succession rate and does succession rate change over time?, and (d) How do plant functional types change during succession? In order to answer these questions, we set 140 4 × 4 m permanent plots in a mobile dune area and monitored vegetation cover and species richness from 1991 to 2011. In time, diversity increased in a logistic manner toward an asymptotic value once vegetation cover surpassed 60 %. Species richness increased in a humped-back shape, also reaching a maximum peak at 60 % vegetation cover. The succession rate of diversity was measured by the Euclidean distance, and showed a significant humped-back relation, meaning that it was slower in early and late successional stages. The study supports the intermediate disturbance theory. The conservation of coastal dunes vegetation should focus on all, species-poor and species-rich habitats that help to maintain the ecological integrity of these ecosystems. The understanding of community dynamics and diversity patterns becomes an essential component of coastal dune management and conservation.
KeywordsChamaecrista Dune stabilization Hump-backed pattern Mexico Primary succession Schizachyrium Succession rate
Coastal dunes occur in many latitudes throughout the world and a wide variety of biomes can develop in these habitats, such as tropical and temperate forests, grasslands, thickets, and typical beach and dune vegetation, tolerant to salt spray, and substrate mobility (Martínez et al. 2004). Besides the biological diversity, coastal dunes are also diverse in terms of landforms, which include different types of dunes (parabolic, barkhans, dune fields, blowouts, foredunes as well as diverse vegetation cover, ranging from no vegetation (mobile dunes) to completely vegetated systems (stabilized dunes). Because of their ample distribution and environmental heterogeneity (Hesp 2000), these ecosystems contain a high ecological diversity and variability of species composition (van der Maarel 1993, 1994). The vegetation patterns and processes of coastal dunes have been of interest to ecologists for more than a century. In fact, as early as 1899, Henry Cowles recognized the basic idea that plant communities of different ages in sand dunes reflected how the communities changed over time.
In addition to the environmental heterogeneity of coastal dunes in terms of local (topographic) and regional (different latitudes) spatial scales, dune heterogeneity is often increased by disturbances which open patches for plant colonization after vegetation is removed and new niches are created where primary succession can be set to its initial stages repeatedly (Hesp and Martínez 2007). Large-scale disturbances that lead to intense sand mobility include severe storms (such as hurricanes) or the transgression of mobile sand from nearby sand deposits that can be carried by ocean currents or the wind action. Thus, the environmental heterogeneity, recurring disturbance events and species interactions result in different patterns and rates of succession. It is also possible to state that temporal and spatial dynamics on sand dunes is an intrinsic and inseparable feature of its ecological integrity.
In the same weather regime, and even in the same location, coastal dunes may be mobile or stabilized and may be located adjacent to each other increasing thus local environmental heterogeneity. Mobile dunes become stabilized (or fixed) when substrate mobility declines because wind speed decreases; because sand supply is depleted; or because vegetation cover increases owing to natural succession, climate change, or human intervention (Yizhaq et al. 2007; Tsoar 2005). Vegetation on mobile dunes remains in its earlier successional stages when dunes are mobile (Moreno-Casasola 1986). As sand mobility decreases, the successional species replacement gets started.
In general, models of ecological succession predict an increase of vegetation cover, richness, and diversity with increasing age (Odum 1971; Stankeviciute 2006). However, this does not seem to be always the case during succession occurring on coastal dunes, where species richness and diversity sometimes decrease in mature stages of succession (Isermann 2011; Peyrat and Fichtner 2011), probably owing to reduced environmental fluctuations (such as diminished disturbances) (Stankeviciute 2006) and changes in species interactions (Connell and Slatyer 1977) which may lead to the dominance of a few species (Pysek and Richardson 2006; Valéry et al. 2008; Muñoz-Vallés et al. 2011). As a result, and contrary to other ecosystems with a linear increment in richness and diversity, succession in sand dunes would show a quadratic function tendency for these parameters.
Since the early 1900s (but more intensively during the last decades), these widespread and diverse ecosystems have been increasingly exposed to additional disturbances: human activities that include tourism, land use change, and urbanization. These actions lead to substantial reductions of species diversity and the loss of ecosystem services such as protection from storms and hurricanes, scenic beauty, and recreation. In particular, tropical coastal dunes are severely threatened because of popular and hence intensified “sand and sun” tourism (Grunewald 2006; Nordstrom 2008; Mendoza-González et al. 2012). Thus, adequate management planning and restoration actions are necessary to preserve these ecosystems. In order to achieve this, the understanding of community dynamics and diversity patterns become an essential component (Acosta et al. 2009).
This study was performed in a well-preserved coastal dune system located in the central region of the Gulf of Mexico, which is representative of nearby dune systems (Moreno-Casasola et al. 1998; López-Portillo et al. 2011). These dune systems are being increasingly threatened by human encroachment including: urbanization, tourism, and land use change (Mendoza-González et al.2012; López-Portillo et al. 2011). Here, we evaluated changes in species richness and diversity of dune vegetation during a 20-year natural succession sequence. We addressed the following questions in our study: (a) Do species richness and diversity increase over time as predicted by models of ecological succession or do they decrease in a hump-backed manner similar to the observations in temperate coastal dune vegetation? (b) What is the interaction between vegetation cover and diversity and between vegetation cover and species richness during the successional sequence? (c) Is there a relationship between species diversity and succession rate and does succession rate change over time? (d) How do plant functional types change during succession?
The central portion of the state of Veracruz, where La Mancha is located, has a general North–South orientation and includes several headlands created by the Mexican Transvolcanic belt which create a series of lagoonal embayments along the shore (Psuty et al. 2009). Sediment transport alongshore occurs from north to south, from one embayment to the next, bypassing these large headlands. In this manner, sediment transported downdrift supports beach and dune forms and dynamics in the next southbound embayment. Twenty years ago (and long before), the coastal dune system at La Mancha was highly mobile and was the recipient of the shifting sands that were transported downwind from the embayment located directly to the north, by the strong winds that occur mostly during the winter months (from November to April). This transgression of mobile sand created mobile dunes at La Mancha as can be observed from aerial pictures that date from 1960 to 2006 (Fig. 1). The situation has changed over the last few decades: the vegetation cover at La Mancha has increased resulting in a decreased area of bare sand and reduced sediment transportation. From 1980 to 2006, 90 ha of the original dunes have changed to agricultural use (2 %), natural grassland (53 %) and coastal thicket and tropical rain forest in several successional stages (35 %). Only 10 % of the originally mobile dunes remain mobile. That is, the quantity of sand moving across the northern headland has decreased in the last 30 years (Psuty et al. 2009) due to a natural successional process, with no artificial plantings of plant species that stabilize mobile dunes, such as marram grass, Casuarina sp., or others. We do not know the precise reasons that have led to these changes in vegetation cover that have resulted in stabilization. However, it is possible that a decline in grazing animals has played an important role (field observations).
Twenty-one years ago, in 1991, we randomly chose a mobile dune area and placed 140 permanent 4 × 4 m plots. The plots were placed forming a grid over the entire area. This mobile dune represented the earliest successional stage, because it was completely mobile, and represented the first stages of primary succession, since there was no pre-existing vegetation. The study site was located ca. 500 m inland, within a heterogeneous dune system in which mobile dunes were adjacent to stabilized dunes covered by tropical rain forest, coastal thicket, and grassland. The total area monitored was 2,240 m2. Each plot was marked with aluminum stakes placed in all four corners of the quadrat. In each plot, we identified all the plant species present; and we estimated their frequency and abundance (as percent vegetation cover). We only recorded vascular plants since there are no bryophytes and lichens in tropical dunes.
The observations took place once a year, at the end of the rainy season (October), the time when plants reached the highest vegetation cover. Vegetation changes were monitored in these permanent plots during two periods, from 1991 to 2000 and from 2009 to 2011, which enabled us to study the successional sequence during 20 years.
Vegetation cover, species richness, and diversity
Statistical analyses were carried out with the program SigmaStat v.11. All raw data were used to calculate species diversity and rate of vegetation change. Our field data with percent estimations were converted into squared meters covered by each species in each plot and then the mean cover value per plot was calculated. In time, as individual plants grew and expanded, species began to overlap in each plot. In consequence, the total area covered by vegetation increased over time and eventually became larger than the total area of each plot.
pi = the relative proportion of every species i in terms of total plant cover (i.e., the relative abundance of species i): ci/C.
ci = relative cover of species i
C = total cover of all individuals of all species
We then performed regression analyses between percent vegetation cover versus species diversity, and between percent vegetation cover versus species richness to test for significant relationships between these variables. In both cases, we used mean values for the regression analyses. The regression models and significance of these correlations were tested as described in each case.
Based on the literature (Moreno-Casasola et al. 1998; Moreno-Casasola and Espejel 1986; Martínez et al. 2001), we distinguished three vegetation functional groups (pioneer colonizers, grassland, and thickets) that we used to assess changes in community structure over time. Once we grouped all species into these three functional types, we added the vegetation cover (in m2) per plot per species for each monitoring date, and then proceeded to calculate the percent cover per functional type by adding the cover values estimated for each species in each functional group.
Vegetation cover, species richness, and diversity
Species richness also increased over time, reaching its maximum value at around ten species per plot in 2009 (Fig. 2b). Overall species richness increased from 14 (in 1991) to a maximum of 30 in 2000. In the last 3 years, total species richness in the monitored area has fluctuated between 24 and 25 species. Similar to the above, mean diversity per plot increased over time too, except in 2011 when it declined slightly (Fig. 2c).
Plant functional types
These trends began to change gradually 5 years after dune colonization started (1994–1998), when grassland species colonized and expanded in the area (Fig. 5). At the beginning of this period, the relative cover of C. chamaecristoides and the perennial grass Schizachyriumscoparium fluctuated around 30–40 %. In the late nineties, grassland species (S. scoparium and Commelina erecta) expanded and became dominant while the relative cover of early colonizers (C. chamaecristoides and P. lindeii) declined markedly.
Finally, in the later successional stages, species typical of thickets have just colonized and are slowly expanding at the site. In the last 3 years, the endemic colonizers were either largely reduced (C. chamaecristoides) or locally extinct (P. lindenii), and grassland species seemed to be declining slightly (Figs. 5 and 6). The colonization by species typical of late successional stages (coastal thickets), such as Opuntia stricta, Randia laetevirens, and Trixis innula took place in the last couple of years, although obviously, at very low frequency and cover values. The vegetation cover of these three species seemed to be increasing gradually (Figs. 5 and 6).
Vegetation cover, species richness, and diversity
This study shows the processes that take place during mobile dune stabilization in tropical latitudes by means of increased vegetation cover from 5 to 95 %. It also describes changes in species richness and diversity during a 20-year natural succession sequence. We found that vegetation cover, species richness, and diversity increased over time, as predicted by successional theory (Odum 1971; Stankeviciute 2006). The variability in mean diversity per plot was much higher in the first years of observation and decreased in the later, because of the colonization process during primary succession. Initially, we had plots with bare sand adjacent to plots that were already being colonized, and thus, quadrats were very different from each other, because of this patchy way of colonization. In time, as bare sand colonization continued, quadrats became more homogeneous as they were all covered by vegetation while bare sand patches disappeared. As the successional sequence continued, all plots were colonized and thus, variability between them decreased because fewer and fewer remained with no vegetation at all. Gradually, plots became more similar to each other. However, considering the whole dune, diversity increased in a logistic way toward an asymptotic value once vegetation cover surpassed 60 %, while species richness increased in a humped-back manner also reaching a maximum peak at 60 % vegetation cover, after which it decreased to lower values similar to early successional stages.
Logistic or monotonic increments in richness and diversity toward a constant or presumed stable state have mostly been observed in mature forests (Margalef 1963; Odum 1969), and also during primary succession after volcanic eruptions (Aplet and Vitousek 1994). More recently, these trends were also found in temperate coastal dunes (Lichter 1998; Isermann 2011; Peyrat and Fichtner 2011). In concordance with our study, in the coastal dunes of Lake Michigan (USA), Lichter (1998) reported that vegetation percent cover varied in a humped-back manner with dune age, and this applied to different functional groups: beach-grass; shrub bunchgrass; and conifers. Similar humped-back relationships were observed by Isermann (2011) and Peyrat and Fichtner (2011) for changes in species diversity and for species richness and diversity, respectively, during primary succession in coastal dunes in Germany and the Baltic coast. In all cases, mid-successional stages showed the highest diversity and species richness values.
In brief, this and other studies (Morrison and Yarranton 1973; Lichter 1998; Vestergaard 2006; Isermann 2011; Peyrat and Fichtner 2011) provide evidence of a general decline in species richness in primary succession on coastal dunes, after initial increase. What are the processes laying beneath these humped-back patterns? Changes in plant species diversity and community composition can be explained by understanding the environmental constraints that potentially influence primary succession and regulate species diversity (Lichter 2000). In coastal dunes, initial succession stages are relatively homogeneous with generally uniform harsh environmental conditions: intense substrate mobility, extreme temperature fluctuations, low nutrient contents, and reduced water availability. Only a few highly specialized species known as psammophytes (C. chamaecristoides, P. lindenii, P. saturejoides and C. punctatus in our case) are able to colonize these environments and support the harsh conditions) (Martínez et al. 2001). In consequence, during the colonization stages of primary succession on coastal dunes most species are tolerant to burial (the pioneers) and very few late colonizer species are present because they are unable to grow in these harsh conditions (Maun 2009). Species–poor vegetation stands are formed with an abundance of early pioneer species. These early successional species (especially C. chamaecristoides) facilitate the colonization of later-successional species by stabilizing the dune surface and by adding nutrients to the incipient soil, and increasing soil moisture (Martínez 2003; Martínez et al. 2001). Colonization by other species after the arrival and expansion of colonizers may occur as follows
As environmental conditions change and become less drastic for non-psammophites, species typical of dune grasslands germinate beneath the early pioneers and grow and reproduce rapidly, expanding at a fast rate (Martínez et al. 2001; Psuty et al. 2009). The gaps of bare sand become reduced or disappear and, in consequence, many light-demanding species begin to become locally extinct. The intermediate stages, with milder environmental conditions (Lichter 1998; 2000; Martínez et al. 2001; Martínez 2003) contain reduced numbers of pioneer species and increased numbers of mid-successional species, such as grasses. In green-house experiments, Martínez and Moreno-Casasola (1996) and Valverde and Pisanty (1999) observed that S. scoparium is a fast growing species in the absence of burial by sand. S. scoparium seedlings have high survival rates on the open dune ridges, but they die under forest cover (Lichter 2000). Larger shade-tolerant species, from coastal dune thickets, are not tolerant to the initial harsh environmental conditions but are hypothesized to be superior competitors for light, and thus, colonize the dune when vegetation cover has increased and may outcompete earlier pioneers and grasses by reducing the amount of available light (Tilman 1985; 1990).
The increment in species richness and diversity in mid-successional stages may be related to intermediate levels of soil pH (Isermann 2005; Moreno-Casasola 1982), increased soil N, soil P, and soil moisture and also to decreasing environmental constraints such as sand movement (Lichter 1998). These mid-successional stages are also related to a higher environmental heterogeneity, because they lie between highly mobile dunes with extreme harsh environment, and stabilized dunes, with no sand movement and milder conditions. Species from both habitats may coexist in these intermediate environments. The frequency of natural disturbance (sand movement) in these habitats is intermediate (Martínez et al. 2001) which thus supports the intermediate disturbance theory of Connell (1978). That is, succession seems to be driven by gradual changes in environmental conditions in which different plant traits become favored (Tilman 1985, 1990; Gallego-Fernández and Martínez 2011).
On top of the above, human intervention may be an additional component driving successional changes. For instance, eutrophication and revegetation can accelerate the rate of succession and even modify community composition by favoring some species (such as grasses) and inhibiting or eliminating others (natives) (Olff et al. 1993; Remke et al. 2009). Alternatively, loss of herbivores (either removed cattle or depleted populations of natural herbivores such as rabbits) may also shift the dominance of species and result in grass encroachment, for example (Kooijman and van der Meulen 1996; Boyes et al. 2010). In any case, the rate of succession will always depend on the disturbance agent and intensity as well as the succession triggers.
Similar to species richness, the succession rate in terms of diversity showed a significant humped-back relation with Shannon-Wienner diversity index, meaning that changes in diversity did not occur in a linear but in a log-normal manner, where succession rate was slower in early and late successional stages and faster in mid-successional stages.
The humped-back shape of the succession rate of diversity that we found in our study coincides with recent studies by Isermann (2011) and can be explained by the changes in community structure that have occurred during the 20 years of successional species replacements. These patterns and processes of dune succession at La Mancha are comparable with successional sequences elsewhere, because the vegetation dynamics is similar although vegetation structure and composition differ between locations (van der Maarel 1993, 1994). In our study site and elsewhere, species-poor vegetation stands are typical of mobile (or yellow) dunes. Usually, the most common pioneer species are grasses and herbs (Morrison and Yarranton 1973; Lichter 1998; Miller et al. 2010; Isermann 2011; Peyrat and Fichtner 2011). In our case, pioneer colonizing species are short shrubs such as the endemics C. chamaecristoides and P. lindenii, as well as other widely distributed species such as C. punctatus. These short shrubs are not only tolerant to burial but their growth and vigor are stimulated when plants are covered by sand (Martínez and Moreno-Casasola 1996). Thus, sand dynamics (erosion and accumulation) is one of the most significant environmental variables driving vegetation and succession changes (Moreno-Casasola 1986; Jungerius et al. 1995; Isermann 2011; Martínez et al. 2001; Maun and Perumal 2002). In these circumstances, succession rate is necessarily low.
As the pioneer colonizing species expanded and the environment became ameliorated (Martínez 2003), new species were able to colonize. In temperate dunes, bryophytes, and lichens as well as shrubs are typical of semi-mobile (or gray) dunes (Lichter 2000; Peyrat and Fichtner 2011; Isermann 2011). In the tropical dunes at La Mancha, grasses and creeping species were most abundant in these stages. Independent of the vegetation structure and composition, the rate of succession in this stage was high because of the colonization of mid-successional species, while early colonizers were still present.
In our study site, we observed that the bunch grass S. scoparius has become dominant in the last 3 years (2009–2011), and this coincided with the dates when the rate of succession was lower. Probably, this is because only a few shade-tolerant shrubs were able to colonize beneath the dense cover of this grass. These observations conform to other studies on the effects of native fast-growing clonal grasses on plant species richness and diversity. For example, many European coastal dunes are experiencing the rapid expansion of native tall-grass species resulting in a reduction of species richness by as much as 60 % (Veer and Kooijman 1997; Kooijman et al. 1998; Provoost et al. 2004; van Til and Kooijman 2007; Ketner-Oostra et al. 2006; Remke et al. 2009).
At this stage, two potential future outcomes are possible: a) the late succession tall shrubs expand gradually and then the rate of succession speeds after the elimination of grassland species and expansion of thickets, when shady conditions increase (Lichter 2000); or b) S. scoparius strongly inhibits the arrival of new species and succession then becomes arrested. Our data suggest a grass encroachment phenomenon and this seems to be occurring in many dune systems in the area (Psuty et al. 2009) but this can only be confirmed after continued monitoring for another decade, at least. If grass encroachment continues, substrate mobility will largely be decreased and biodiversity will deplete after the loss of early colonizers.
The stabilization of coastal dunes is the result of the activation of the successional process triggered by either natural or anthropic causes. Thus, monitoring and understanding vegetation dynamics is essential for their protection and conservation at different scales, for supporting sustainable development. This study adds to the growing list of evidence that demonstrates that the highest coastal dune diversity values are largely linked to a heterogeneous environment with different stages of the successional sequence, including early and late seres. Therefore, and in coincidence with Acosta et al. (2009) and Peyrat and Fichtner (2011), the conservation of coastal dunes vegetation should focus on all, species-poor and species-rich habitats that will largely help to maintain community dynamics and species diversity.
This study was partially funded by grant from SEMARNAT-CONACYT (Secretaría de Medio ambiente y Recursos Naturales-Consejo Nacional de Ciencia y Tecnologia) (23669) and two scholarships (LLAM and PFB). We are very grateful to Patricia Moreno-Casasola, Victor Parra-Tabla, and Gabriela Vázquez for their useful comments on early versions of the manuscript. Thanks to Deborah Lithgow for field assistance. Figure 1 was created by R. Landgrave.
- Hesp PA (2000). Coastal sand dunes: form and function. CDVN Technical Bulletin no. 4. Forest Research. p 30Google Scholar
- Lichter J (1998) Primary succession and forest development on coastal Lake Michigan sand dunes. Ecol Monogr 68(4):487–510Google Scholar
- López-Portillo J, Martínez ML, Hesp PA, Hernández-Santana JR, Vásquez-Reyes VM, Gómez-Aguilar LR, Méndez-Linares AP, Jiménez-Orocio OA, Gachuz-Delgado S (2011) Atlas de las costas de Veracruz: manglares y dunas. Secretaría de Educación y Cultura del estado de VeracruzGoogle Scholar
- Martínez ML, Moreno-Casasola P (1996) Effects of burial by sand on seedling growth and survival in six tropical sand dune species. J Coastal Res 12(2):406–419Google Scholar
- Martínez ML, Psuty N, Lubke R (2004) A perspective on coastal dunes. In: Martínez ML, Psuty N (eds) Coastal dunes: ecology and conservation. Springer-Verlag, Berlin, pp 3–10Google Scholar
- Maun MA (2009) The biology of coastal sand dunes. Oxford University Press, OxfordGoogle Scholar
- Mendoza-González G, Martínez ML, Pérez-Maqueo O, Lithgow D, Simonin P (2012) Land use change and its effects on the value of ecosystem services along the coast of the Gulf of Mexico. Ecol Econ Google Scholar
- Moreno-Casasola P (1982) Ecología de la vegetación de dunas costeras: factores físicos. Biotica 7:577–602Google Scholar
- Moreno-Casasola P, Monroy R (2006) Introducción. In: Moreno-Casasola P (ed) Entornos veracruzanos: la costa de La Mancha. Instituto de Ecología, A.C., Xalapa, Veracruz, Mexico, pp 17–23Google Scholar
- Moreno-Casasola P, Espejel I, Castillo S, Castillo-Campos G, Durán R, Pérez-Navarro JJ, León JL, Olmsted I, Trejo-Torres J (1998) Flora de los ambientes arenosos y rocosos de las costas de México. In: Halffter G (ed.) La diversidad biológica de Iberoamérica Vol. II. Acta Zoológica Mexicana, nueva serie. Volumen especial, pp 177–258Google Scholar
- Odum E (1971) Fundamentals of ecology. Saunders, Philadelphia, p 624Google Scholar
- Stankeviciute J (2006) The succession of sand vegetation at the Lithuanian seacoast. Bot. Lit. 12(3):139–156Google Scholar
- Van der Maarel E (1993) Dry coastal ecosystems: polar regions and Europe. (Ecosystems of the World). Elsevier Science, AmsterdamGoogle Scholar
- Van der Maarel (1994) Dry Coastal Ecosystems: Africa America Asia and Oceania (Ecosystems of the World). Elsevier Science, AmsterdamGoogle Scholar
- van Til M, Kooijman (2007) A Rapid improvement of grey dunes after shallow sod cutting. In: Isermann M, Kiehl K (eds.): Restoration of Coastal Ecosystems. Coastline Reports 7 pp. 53–60Google Scholar
- Weisstein EW (2012) Distance. From MathWorld-A Wolfram web resource. URL: http://mathworld.wolfram.com/Distance.html Accessed 26 Mar 2012
- Yizhaq H, Ashkenazy Y, Tsoar H (2007) Why do active and stabilized dunes coexist under the same climatic conditions? doi:10.1103/PhysRevLett.98.188001