Advertisement

Hydrobiologia

, Volume 814, Issue 1, pp 5–17 | Cite as

Interactions between flooding and upland disturbance drives species diversity in large river floodplains

  • Mauricio E. Arias
  • Florian Wittmann
  • Pia Parolin
  • Michael Murray-Hudson
  • Thomas A. Cochrane
MULTIFUNCTIONALITY OF LARGE RIVERS

Abstract

Understanding and predicting vegetation patterns in floodplains are essential for conservation and/or restoration of river floodplains subject to hydrological alterations. We propose a conceptual hydroecological model to explain the disturbance mechanisms driving species diversity across large river floodplains. These ecosystems harbor a unique set of flood-tolerant species different from the surrounding upland vegetation. In elevation gradients across pristine floodplains, the greater the flooding, the fewer the number of plant species. As terrain elevation increases, flood depth and duration decrease and it is more likely that species composition is influenced by external natural or human-driven disturbances. The spatial interaction between the natural flood regime and upland factors creates patterns of disturbance gradients that influence how floodplain vegetation establishes. In regions where upland conditions are subject to strong external disturbances, species diversity peaks at intermediate stages along the disturbance gradient. We demonstrate this concept with observations from the Central Amazon and Pantanal in Brazil, the Mekong’s Tonle Sap in Cambodia, and the Okavango Delta in Botswana. We discuss how this model could be further elaborated and validated to inform management of large river basins under the impact of upstream-induced flood pulse alterations.

Keywords

Floodplain ecology Tropical rivers Plant species diversity Flood pulse concept Intermediate disturbance hypothesis Flood hydrology 

Introduction

River floodplains are complex and productive ecosystems that provide essential services to nature and society. They provide rich habitats and food sources for both terrestrial and aquatic organisms, while supplying drought protection, fisheries, and agricultural grounds to riverine human populations. Overall, the value of wetland ecosystems services around the globe is estimated to be higher than any other inland landscape (De Groot et al., 2012), but the appropriate functioning and provision of floodplain ecosystem services to humans and nature depend on the degree and frequency of connectivity with the upstream river system. These two factors, however, have drastically changed in rivers around the world, primarily as a result of water infrastructure developed to regulate water for hydropower, water supply, and irrigation.

Although water infrastructure development has presumably already resulted in drastic alterations to the hydrology and biota of vast areas of floodplains around the world, the future management of those already disturbed ecosystems and other less disturbed ecosystems will bring multiple challenges. On the one hand, water infrastructure development is accelerating in emerging economies, and future plans will likely impact conservation strategies for those large rivers that are still relatively unregulated (Grill et al., 2015). The changing climate is also bringing modifications to the main drivers of floodplains’ productivity, including extreme hydrological events, rising temperatures, and salt intrusion (Hamilton, 2010; Junk et al., 2013). On the other hand, there are an increasing number of opportunities for large scale hydrological and ecological restoration, as demonstrated by proposed and ongoing programs in Europe (European Commission, 2015) and North America, in particular in the Everglades (Sullivan et al., 2014), Mississippi River (Mitsch & Day, 2006), and Colorado River (Glenn et al., 2013). Whether the focus is conservation, climate change adaptation, or restoration, understanding and predicting long-term ecological patterns and links to the hydrology are essential for the sustainable management of large rivers and their floodplain ecosystems.

The objective of this paper is to propose a conceptual model of how inundation patterns in the annual to decadal time scale controls plant species diversity gradients in large river floodplains and how these may be affected by hydrological alterations. The paper first provides an overview of the current status of large floodplains and how they are affected by river regulation. A brief description of two relevant ecological concepts that underlay this paper’s argument is then provided. Following this, the conceptual model proposed is described. Case studies of four large floodplains on three continents—the Mekong, the Amazon, Pantanal, and Okavango—are then reviewed in light of species richness patterns observed in the field and how these observations relate to the conceptual model. The paper then concludes with a statement of opportunities for future research. The concepts reviewed and proposed throughout this paper aim to provide a crucial step in developing a practical tool that can be used to support management and restoration of large river floodplains around the world.

Current status and future perspective of large river floodplains

Nearly half of the world’s large rivers that once had extensive floodplains are now regulated for hydropower, irrigation, or flood control (Nilsson et al., 2005; Lehner et al., 2011). One way to exemplify the magnitude of alteration of large floodplains with respect to river alterations is by mapping the global extent of inundation patterns with respect to water development infrastructure (Fig. 1). Looking first at the distribution of floodplain wetlands and forests around the world as mapped by Lehner & Döll (2004), it is clear that most of the seasonally inundated areas are part of major river basins within the tropics like the Orinoco, Paraguay, and the Amazon in South America; Congo and Zambezi in Africa; and Indus, Ganges, Mekong, and Yangtze in Asia. These large rivers, however, have drastic differences in their degree of alteration caused by water infrastructure development as highlighted by the map of river dam regulation by Lehner et al. (2011). For instance, there are very high levels of regulation in the Mississippi and Indus-Ganges basins, where intensive irrigation agriculture is dominant. The Mekong and Yangtze rivers in Southeast Asia, with large hydropower dams in their headwaters, have comparatively mild levels of regulation. The mainstem of the Amazon and Congo—the rivers with the largest discharge—have had little regulation in past decades, despite recent efforts and future plans to construct large dams in both basins (Winemiller et al., 2016). These differences in degree of river alteration are also a direct reflection of the ecological and biodiversity status of their floodplains, which is certainly most impoverished in heavily altered rivers like the Mississippi and best preserved in the least regulated Amazon and Congo. The remaining unregulated, contiguous, and undisturbed large river floodplains are small in number, but large in area and biodiversity of both terrestrial and aquatic species. We expect, therefore, that studying hydrological-ecological patterns in a variety of large rivers would provide important information to facilitate conservation of pristine floodplains as well as adaptation and restoration of those that are more heavily altered.
Fig. 1

Global distribution of large river floodplains and flooded forests (blue shade) and dams (black dots) with data from Lehner & Döll (2004) and (Lehner & Döll, 2004; Lehner et al., 2011). Basin case studies in orange

Hydrological controls in floodplain species: convergence of ecological concepts

In order to conceptualize a link between hydrological controls and diversity patterns in large river floodplains, it is important to provide the theoretical framework underlying this relationship. A general model of spatial biocomplexity across river networks has been previously proposed (the riverine ecosystem synthesis; Thorp et al., 2006), and the intent of our conceptual model is to synthesize and harmonize knowledge from this and previous concepts of critical importance to the understanding and management of large river floodplains. In particular, there are two widely known concepts in freshwater ecology that are highly relevant to the topic presented in this paper: the intermediate disturbance hypothesis (IDH; Connell, 1978) and the flood pulse concept (FPC; Junk et al., 1989). The following paragraphs give a short overview of these concepts and how they relate to the conceptual hydroecological model presented in this paper.

Intermediate disturbance hypothesis

The IDH first proposed that maximum levels of biological diversity occur at intermediate levels of disturbance with respect to time and/or space. Too much disturbance allows only for the establishment of a few pioneering species, whereas low disturbance facilitates the eventual dominance of a selected number of climax species. Hence, intermediate stages of disturbance offer a transitional phase/niche in which pioneering, intermediate, and climax species may co-exist. Connell (1978) first postulated this hypothesis for rainforest trees and coral reefs, but thereafter the IDH has been proven (and disproven) for a wide range of ecosystems (Petraitis et al., 1989; Mackey & Currie, 2001; Molino & Sabatier, 2001; Tanentzap et al., 2013; Jardine et al., 2015). It is important to recognize that the IDH only represents a set of similar phenomena that arise from different co-existing mechanisms (Roxburgh et al., 2004), but it does not explain the underlying mechanisms themselves.

Floodplains are among the list of ecosystems in which the IDH has been studied. Floodplains (which may be described as riparian wetlands) are at interphase zones between terrestrial and aquatic ecosystems (Naiman & Décamps, 1997), thus harboring species from both types of ecosystems. In addition, floodplains may harbor a different set of species altogether that may or may not be greater in number than their terrestrial counterparts (Sabo et al., 2005). Clear unimodal patterns of maximum number of species as predicted by the IDH have been observed as a function of flooding frequency (Pollock et al., 1998), and Tanentzap et al. (2013) stated that diversity patterns in wetlands are a result of differences in species’ niches rather than demographic stochasticity. Contrary to IDH predictions, Zelnik & Carni (2008) found that species richness increased linearly as a terrain elevation along a wetland moisture gradient surrounded by a landscape with low land use intensity.

The flood pulse concept

The FPC states that the regularity and predictability of the seasonal delivery of river water onto floodplains is the main driver of biogeochemical cycles, habitat characteristics, and species distribution in large river floodplains (Junk et al., 1989). The FPC was primarily based on research in the Central Amazon floodplains, where a range of ecosystem components including chemical cycles (Kern & Darwich, 1997; Wassmann & Martius, 1997; Weber, 1997), flora (Junk & Piedade, 1997; Ferreira & Stohlgren, 1999; Wittmann et al., 2011), and fauna (Petry et al., 2003; Lobón-Cerviá et al., 2015) were analyzed over the last few decades. The basic assumption of this concept is that the regular and predictable flood pulse is the main ecological factor influencing all organisms living in that ecosystem. Indeed, recent research has quantified and demonstrated that flood predictability (or rhythmicity) across river systems determines aquatic species richness and plant productivity (Jardine et al., 2015). Flooding regularity and predictability allow the evolution of adaptations and of highly diverse biota. They maintain genetic and species diversity in the floodplain ecosystem. Parolin & Wittmann (2010) suggested that the FPC could also apply to other (sub-)tropical floodplains, and observations in support of this concept have been documented in the Orinoco (Lewis et al., 2000), Mekong (Arias et al., 2013; Holtgrieve et al., 2013), Okavango (Davidson et al., 2012), and Australia’s wet–dry tropics (Warfe et al., 2011; Jardine et al., 2015). Clearly, the simplicity in which the FPC describes floodplain processes has been criticized (Zurbrügg et al., 2012), in particular with respect to the catchment-river-floodplain water exchange, demonstrated to occur in complex ways involving local catchment inputs, secondary channels, and banks (Lesack & Melack, 1995; Zurbrügg et al., 2012; Rudorff et al., 2014). While it is indeed important to underpin hydrological and biological mechanisms in floodplains, the FPC was coined as a conceptual model that summarizes the overall result of all these processes (Bayley, 1995).

A conceptual model of species diversity for river floodplains

We hypothesize that both concepts described above bring important aspects to the spatial distribution of plant species in river floodplains. The FPC suggests that those floodplains areas with an intermediate and predictable flooding regime exhibit the highest biodiversity (Junk & Wantzen, 2004), suggesting that there is an underlying direct connection with the IDH as pointed out by Jardine et al. (2015). Based on these two ecological concepts, we propose a conceptual model in which species diversity varies in floodplain landscapes as a function of flooding patterns and the land use/land cover conditions that dominate the surrounding uplands.

Large river floodplains typically occur along low terrain elevation gradients. Because of the seasonality in the flood pulse, sub-meter elevation differences translate into a strong flooding gradient. As flooding intensity and frequency diminish, conditions become more favorable for upland disturbances such as agriculture, drought, and fire; thus, one could say that in the opposite direction of flooding, there is an upland disturbance gradient (Fig. 2). Although unregulated flooding is a natural phenomenon in large river floodplains, the flooding gradient also acts as a disturbance mechanism to those species that cannot tolerate water logging.
Fig. 2

Floodplain terrain elevation profile. Flooding and upland gradients in opposite directions (upper frame). Hypothetical response of floodplain species diversity (lower frame)

As a result of the flooding gradient, floodplains harbor a unique set of species with different degrees of tolerance to flooding, and which are typically different from species from the surrounding uplands. In an elevation gradient across a pristine floodplain, it is expected that the greater the flooding, the fewer the number of plant species that can tolerate such conditions. As the terrain elevation increases, the more favorable the conditions for rooted vegetation are, and therefore, the number of species is likely to increase to the extent at which upland disturbances become an important factor. If seasonal flooding becomes marginal due to upstream factors (e.g., water resources development), it is more likely that species diversity can be influenced by external (upland) disturbances. These can not only be anthropogenic (e.g., agriculture, deforestation, etc.), but they can also be natural (e.g., drought and fires). The spatial interactions between the natural flood regime and upland factors create different patterns of disturbance gradients that influence how vegetation establishes in a floodplain. In floodplains that are surrounded by uplands with optimal conditions for vegetation growth and little external disturbance—as it is in the case of the Central Amazon—species diversity responds inversely proportional to flooding. However, in regions where upland conditions are subject to strong external disturbances, species diversity peaks at an intermediate stage along the disturbance gradient (Fig. 2).

River floodplain case studies

In order to illustrate the concept presented in this paper, we selected four large floodplain systems across the tropics in South America, Africa, and Asia where species diversity distribution data with respect to flooding have been published (see Fig. 1 for basin locations). These four case study systems experience distinct seasonal flood pulses that have had historically similar low levels of flow regulation. The surrounding uplands, however, represent very different environmental conditions and disturbance types that could partially explain variations in plant species diversity gradients among these floodplains. Species diversity patterns presented below are based on the field plot inventories, but in all four cases the data represent the summary of observations across the entire study sites.

The Tonle Sap: Mekong’s floodplain within the rice paddies

The Mekong is Southeast Asia’s largest river basin, covering an area of approximately 795,000 km2. The floodplains cover over 41,000 km2, distributed primarily between the Mekong Delta in Vietnam and the Tonle Sap in Cambodia. The latter is a lake–floodplain system that exchanges a maximum discharge of 10,000 m3/s with the Mekong via the Tonle Sap River. Overall, water levels in the Tonle Sap floodplain fluctuate by an average of 8 m annually, inundating an area over 13,000 km2 for an average of 4 months per year (Arias, 2013).

Recent field observations demonstrate that habitats, species diversity, and other vegetation characteristics in the Tonle Sap floodplain vary with respect to elevation, inundation patterns, and land use (Fig. 3; Arias et al., 2013). Immediately upland from the permanent lake and aquatic grasslands, there is a zone of tall and open canopy that is commonly referred to as gallery forest. Such habitat is inundated around 9 months per year (Arias et al., 2012) and it is dominated by Barringtonia acutangula Gaertn., among a few other species of woody vegetation. Further upland, a greater number of shrub and grass species (with shorter but denser canopy) are present in a zone that remains flooded for 6–8 months. The lower canopy in this part of the floodplain may occur as a result of the greater drought stress experienced during the short period without inundation (McDonald et al., 1997). The shrub species Vitex holoadenon D. Dop (Verbenaceae) and Mimosa pigra L. (Fabaceae) are abundant in these habitats. It is important to note that this last species is an aggressive invasive that has been observed primarily on sites cleared in recent years (Arias et al., 2013) and it is hypothesized that increased disturbance has enhanced the spreading of this invasive. As one reaches areas inundated 3–5 months per year, species diversity peaks as a result of two factors. First, a diverse set of natural and agricultural habitats co-exist in this zone. Less duration and depth of inundation means greater opportunities for cropping, and indeed, traditional varieties of floating rice have been used in the past, but their presence has been declining, since the early 1980s as a result of civil war and transition to conventional rice varieties that cannot withstand deep flooding for long periods of time (Nesbitt, 1997; Sarkkula et al., 2003). In addition to habitat diversity, the non-agricultural area of this zone has the greatest diversity among all floodplain habitats of this ecosystem. This increase in species diversity with decreasing flooding intensity is what is expected in natural wetland habitats and can be explained by the increasing favorable conditions for terrestrial plant growth. Following this intermediate zone, however, species diversity declines again; the outer zone of the floodplain remains inundated for a month or less annually, and it might not even receive any flooding from the Mekong during drier years. Thus, this outer zone is dominated by conventional rice paddies that depend on controlled irrigation.
Fig. 3

Plant species alpha-diversity (Shannon Index) in the Tonle Sap. Data from 100 m2 plots as described in Arias et al. (2013). Standing biomass per species used as surrogate of proportional abundance. Photographs represent vegetation type along the flood gradient

The Okavango Delta: an Oasis in the Kalahari Desert

The Okavango river basin straddles three southern African countries and extends 1100 km from upper catchment to the terminal Delta. This basin extends across approximately 120,000 km2 of sub-humid to semi-arid uplands, with a rainfall gradient from 1400 mm in the north to 450 mm in the south. Rainfall occurs between September and April, peaking in December to February, sending annual pulsed flows down the river system, through Namibia into the fault bounded alluvial fan which forms the Okavango Delta in Botswana.

The average annual inflow to the Delta is approximately 10 km3 (Wilson & Dinçer, 1976; Porter & Muzila, 1989). Under current climatological conditions, the maximum extent of inundation varies from 8000 to 12,000 km2, of which about 6000 km2 is perennially inundated (Gumbricht et al., 2004a, b). Each distributary system across the alluvial fan exhibits slightly different flood behaviors; the westerly and central Thaoge and Jao distributaries, which receive primarily overbank flow, are characterized by fluctuations in water level of 1.5–2 m, while the easterly Moanatshira, Kwhai, and Mogogelo systems are fed by base flow, and the amplitude of fluctuation is less than 0.6 m (Porter & Muzila, 1989).

In general, local elevation differences between channel bed or floodplain floor, and island high-water line are in the order of 1–2 m (Gumbricht et al., 2004a), and zoning of herbaceous vegetation in the floodplains reflects this topographic gradient. However, interannual and multi-decadal variations in hydroperiod are such that vegetation zones are not spatially stable. The Delta falls into the Zambezian Flooded Grasslands Ecoregion (Snowy Mountains Engineering Corporation, 1989), characterized by a paucity of flood-tolerant woody species; consequently, floodplains are dominated by herbaceous plants, primarily grasses and sedges, but with many herbs and forbs. Within floodplain catenas, there is a general trend for the upper, shorter hydroperiod zones to be dominated by grasses, while the proportion of sedges increases down-slope toward the longer hydroperiod zones (Smith, 1976; Murray-Hudson et al., 2011). Perennially flooded areas are characterized by hydrophytes, and generally exhibit lower diversity than areas which experience a dry period at the low point of each pulse. The invasive Salvinia molesta has been found to compete well with other hydrophytes in the predominantly nitrogen-limited environment, but an efficient bio-control system for this invasive is in place in the Delta. Work done in the seasonal floodplains (Murray-Hudson et al., 2015) indicates that peaks in richness and diversity occur at medium (6 month) and high (10 month) mean annual flood duration, and the dry and wet ends of the duration range are characterized by lower numbers of species (Fig. 4). The lowest numbers of species and plant diversity occur in perennially inundated areas and in areas with a mean duration of less than 6 months. Murray-Hudson et al. (2015) found that at higher flood frequencies, depth of flooding comes into play as a driver of composition.
Fig. 4

Alpha-diversity (Shannon Index) of herbaceous species along the flood gradient of the Okavango Delta. Summary of observations in number of species per 1 m2 quadrant. Photographs represent the vegetation along the flood gradient. Data from Murray-Hudson (2009)

The Pantanal: high biodiversity between inundation and drought

With an area of approximately 160,000 km2, the Pantanal is one of the largest seasonal wetlands on Earth. Located within the neotropical savanna-belt (Cerrado), it is seasonally flooded by the upper Paraguay River and several of its tributaries. The precipitation regime in the region is clearly seasonal, with a dry period lasting from April/May to September/October, while the rainy period contributes approximately 70% of total annual precipitation (1090–1250 mm). This leads to seasonal water-level changes of the upper Paraguay River with a mean amplitude of 3.1 ± 0.9 m (data from 1988 to 2007, Wittmann et al., 2008).

The landscape of the Pantanal consists of a patchwork of seasonally inundated and non-inundated habitats (Zeilhofer & Schessl, 2000). Non-inundated habitats include isolated granitic outcrops and paleo-levees that were mostly formed as river terraces during Pleistocene and Holocene interglacial periods. These elevations are important refugia for savanna species which are characterized by low tolerance against seasonal inundation.

The vegetation of the Pantanal consists of semi-deciduous forests upon non-inundated levees and seasonally inundated riparian forests along the Paraguay River and most of its tributaries. Seasonally inundated savanna vegetation predominates between these formations, varying from open grasslands to shrub and tree savannas, the latter often characterized by mono-dominant formations of shrub or tree species (Prance & Schaller, 1982).

Seasonal inundated riparian forests are the most species-rich forest types within the Pantanal, as they comprise approximately 50% of the 750 registered woody species within the region (Pott & Pott, 1994; Nunes da Cunha & Junk, 2001). Tree species richness and composition in riparian forests of the Pantanal are influenced by small differences in topography and seasonal inundation (Fig. 5). Tree species might be inundated by up to three meters in height (Damasceno-Junior et al., 2005), which corresponds to a mean flood duration of 220 days year−1 (greater than 7 months). In riparian forests of the Miranda River, Southern Pantanal, mean flood duration of trees varies from 0 to 160 days year−1 within a topographic range of only 0.5 m (Wittmann et al., 2008). At highly inundated sites, forests are composed of few flood-tolerant species, such as Alchornea castaneifolia Willd. A. Juss. (Euphorbiaceae), Triplaris gardneriana Wedd. (Polygonaceae), and Crataeva tapia L. (Capparaceae). With decreasing inundation, tree species richness increases locally, and is highest in the transition between seasonally flooded and non-flooded habitats, with up to 45 species ha−1 (≥10 cm diameter at breast height; Wittmann et al., 2008). Maximum tree species richness in non-inundated habitats of the Pantanal ranges from 24 to 36 species ha−1 (Ratter et al., 1988; Dubs, 1992), thus comparatively low. This reflects the gradient of increasing disturbance in upland forest, notably the effect of seasonal droughts that eliminate the occurrence of moisture-sensitive tree species, an increased frequency of natural and man-made fires, as well as the impact of deforestation, agriculture, and cattle ranching, the main economic activity in the region.
Fig. 5

Alpha-diversity (Shannon Index) of trees in a riparian forest of the lower Miranda River, Southern Pantanal, plotted against inundation period. Each dot represents a forest plot with an area of 25 × 25 m (625 m2), where all trees ≥10 cm diameter at breast height were recorded. Tree species diversity per plot ranged from 6 to 13 tree species, all 16 plots combined had a total diversity of 45 species. Mean flood duration was estimated from daily flood-level data of the Ladário gage, Paraguay River, during the period 1988–2006. Figure adapted from Wittmann et al. (2008)

Central Amazon Floodplains: an exception to the model, but for how long?

The Amazon basin comprises an area of 6915,000 km2 shared among seven countries on the South American continent. 11–12% of the Amazon basin (or approximately 800,000 km2) is comprised of floodable areas, not including intermittently flooded moderate to small rivers (Melack & Hess, 2011). The Amazon River has a maximum discharge of 278,000 m3/s (as measured in Óbidos in the Central Amazon), and water-level fluctuations are 4–10 m in the forested floodplains (Junk et al., 1989, 2011). Flooding seasonality is regular, with a high-water peak in June and a low water trough in October. This latter month is congruent with the period of low precipitation and thus leads to occasional vegetation drought stress (Parolin et al., 2010). The dominant vegetation in the floodplain is highly diverse forest, which experiences on average 140 days year−1 of flooding (Wittmann et al., 2011). Areas lying lower in the flood gradient are dominated by highly productive grasses and macrophytes and experience maximum flood duration of 270 days year−1 (Junk & Piedade, 1997). The várzea forests, which occur in nutrient-rich white-water floodplains, have different associations of tree species that form a clear zonation (Wittmann et al., 2004). High várzea forests adjacent to the upland forests are among the most diverse floodplain forests on the planet (Wittmann et al., 2006). The seasonal flood pulse associated with the hydrogeomorphological dynamism of várzea floodplains implies that the ecological factors important for the establishment and distribution of tree species in várzea forests are relatively constant over thousands of kilometers along the river courses (Wittmann et al., 2006). Species colonization is driven by the selection pressure provided by flooding and the associated hydrogeomorphological factors (Wittmann et al., 2011). Mechanisms of competition in specialist (or highly flood-adapted), tree species lead to a point of no return in the evolution of adaptations to the seasonal inundation, implying that competition of these species is successful only in flooded environments. This tree species colonization concept states that the ecotype thus became endemic to the floodplains. More than 10% of the 685 most common central Amazonian várzea tree species are endemic (Wittmann et al., 2013). With this high degree of endemism, Amazonian large river floodplains represent a unique status among floodplain forests, which tend to be scarce in endemic species and are generally composed of widely distributed species typical in lowland rainforests and savannas (Wittmann, 2012). Beyond the floodplains, the Amazon uplands offer optimal environmental conditions that lead to remarkably large diversity. A much greater—and different—number of non-flood-tolerant species surround the floodplains, causing the gradient of species diversity to decrease linearly along the flood gradient (Fig. 6). However, land use practices and the changing climate are promoting deforestation and drought throughout the Amazon (Malhi et al., 2008; Coe et al., 2013), and therefore it is possible that diversity gradients in the Amazon floodplains may transform to more similar patterns as those observed in other floodplains where upland disturbances are more prominent.
Fig. 6

Alpha-diversity (Fisher’ coefficient) of 25 forest plots with an area of 1 ha within Amazonian white-water (várzea) forests plotted against mean flood height. Forest inventories were performed by different authors, as described by Wittmann et al. (2006). Tree species diversity per plot ranged from 17 to 109 tree species. Flood-level data derive from daily water-level measurements of the Manaus gage at the confluence of the Negro and Amazon Rivers during the period 1903–2004. Figure adapted from Wittmann et al. (2006)

Opportunities for future research

The concept and case studies presented above are initial indications of what could be a model that applies to a much wider range of large river floodplains worldwide. There is, however, much more that we need to learn about these ecosystems before a comprehensive view of their contemporary and future status can be assessed. First, we propose that studies across multiple river basins are carried out comparing mostly pristine systems to systems affected by extensive water infrastructure development. This would facilitate the standardization of data collection methodologies, the increase in number of case studies available to better understand key factors influencing species diversity in large river floodplains, and it would also provide a means to test our hypothesis. The use of remote sensing technology, for instance, is a promising tool that has been used to study tree species distribution and diversity (Wittmann et al., 2002; Saatchi et al., 2008; Betbeder et al., 2014) and could facilitate testing our concept in basins that have not been surveyed in great detail. Also, it could provide a way to standardize data collection, spatial resolution, and time frames. Satellite-based remote sensing products have been widely used for assessing inundation patterns in the case studies described in this paper (e.g., Hamilton et al., 2002; Van Trung et al., 2012; Gumbricht et al., 2004b), and links between these patterns and vegetation patterns are evident (Fig. 7; Wittmann et al., 2002; Arias et al., 2012; Murray-Hudson et al., 2015). Assuming that there is a good relationship between plant species composition and habitat characteristics detected through satellite sensors, the use of this technology to further elaborate our concept appears to be a promising alternative.
Fig. 7

Remote sensing observations can be used as a surrogate of gradients driving species diversity; a MODIS Satellite-derived Normalized Difference Vegetation Index (NDVI) from during the dry season in the Tonle Sap, where dark blue indicates open water and green emergent vegetation; b Land use/Land cover classification for the same region, follow very similar patterns as the NDVI. Land use/Land cover data from Cambodia’s Ministry of Agriculture, Forestry, and Fisheries. Both frames created by the authors

Once sufficient evidence is gained about how flooding and upland disturbances affect species diversity in floodplains, an important step to follow is the development of a numerical model. Such model would be a useful tool in projecting floodplain changes as a result of hydrological alterations (or restoration), human use, and changes in climate. This model needs to be a practical tool to be used across systems, primarily driven by data that are globally available. What is probably most innovative and challenging is that such a model ought to integrate human land use disturbance as a mechanism that is interrelated to hydrological processes. Moreover, the model should lead to spatially explicit results that are useful to multiple stakeholders, and its inputs should be compatible with hydrological models that are commonly used for basin and floodplain management. A good example of what a baseline for this model could exist for the Everglades (Foti et al., 2012), but much more data collection and model development need to be carried out before a similar tool can be generalized for floodplains worldwide.

Conclusions

This paper argues that the spatial distribution of species diversity along inundation gradients in large river floodplains can be explained by the interaction between two critical environmental drivers: flooding and upland disturbances. We described a conceptual model of how these two drivers interact in unregulated floodplains, and how the frequency and magnitude of flooding can actually shape the extent and type of land use and cover that takes place in a floodplain. In floodplains where there is little external upland disturbance, species diversity responds inversely proportional to flooding. In regions where upland conditions are subject to strong external disturbances, however, species diversity peaks at intermediate stages along the disturbance gradient. Our model links elements from two well-established ecological concepts, the Intermediate Disturbance Hypothesis (Connell, 1978) and the Flood Pulse Concept (Junk et al., 1989), which appear to have an excellent synergy when explaining patterns of species diversity in floodplains.

The concept described in this paper applies to a large extent of the wetland biome, concentrated primarily in large river floodplains within the tropics (Fig. 1). In order to illustrate this concept, we selected four case studies, for which there are published survey data on species diversity. Overall, we found that our conceptual model fits well with the observations in those systems like the Tonle Sap, Okavango, and the Pantanal where agriculture and drought are common in the outer inundation boundaries, creating a distinct species diversity peak in the middle sections of the floodplain and decreasing both toward drier and wetter conditions. Species diversity in the Central Amazon, however, declines proportionally to flooding, primarily driven by the near-optimal conditions for vegetation growth in the Amazon uplands. Since deforestation and drought are serious threats to the Amazon floodplains, species diversity patterns could shift in the future to a state similar to that found in the other case studies. Such shifts could ultimately bring serious consequences to the Amazon floodplains, compromising the functions and services that these ecosystems provide.

It is not only the Amazon, Tonle Sap, Pantanal, and Okavango which face an uncertain future; these are only a selected sample of large floodplains that are increasingly changing as a result of water infrastructure development and climate change. There is a serious need for practical management tools that make use of the little existing information in combination with remote sensing data and mathematical models, and we hope that this paper provides a step forward in scoping such tools that can be used for adaptation and restoration of these remarkably important ecosystems.

Notes

Acknowledgments

This manuscript was completed while M. E. Arias was a Giorgio Ruffolo Fellow in the Sustainability Science Program at Harvard University. Support from Italy’s Ministry for Environment, Land and Sea is gratefully acknowledged. Comments from two reviewers were very helpful in improving the original manuscript.

References

  1. Arias, M.E., 2013. Impacts of hydrological alterations in the Mekong Basin to the Tonle Sap ecosystem (PhD thesis). University of Canterbury, Christchurch.Google Scholar
  2. Arias, M. E., T. A. Cochrane, M. Kummu, T. J. Killeen, T. Piman & B. S. Caruso, 2012. Quantifying changes in flooding and habitats in the Tonle Sap Lake (Cambodia) caused by water infrastructure development and climate change in the Mekong Basin. Journal of Environmental Management 112: 53–66.CrossRefPubMedGoogle Scholar
  3. Arias, M. E., T. A. Cochrane, D. Norton, T. J. Killeen & P. Khon, 2013. The flood pulse as the underlying driver of vegetation in the largest wetland and fishery of the Mekong Basin. AMBIO 42: 864–876.CrossRefPubMedPubMedCentralGoogle Scholar
  4. Bayley, P. B., 1995. Understanding large river-floodplain ecosystems. BioScience 45: 153–158.CrossRefGoogle Scholar
  5. Betbeder, J., V. Gond, F. Frappart, N. N. Baghdadi, G. Briant & E. Batholome, 2014. Mapping of central Africa forested wetlands using remote sensing. IEEE Journal of selected topics in Applied Earth Observation and Remote Sensing 7: 532–542.CrossRefGoogle Scholar
  6. Coe, M. T., T. R. Marthews, M. H. Costa, D. R. Galbraith, N. L. Greenglass, H. M. A. Imbuzeiro, N. M. Levine, Y. Malhi, P. R. Moorcroft, M. N. Muza, T. L. Powell, S. R. Saleska, L. A. Solorzano & J. Wang, 2013. Deforestation and climate feedbacks threaten the ecological integrity of south–southeastern Amazonia. Philosophical Transactions of the Royal Society B: Biological Sciences 368(1619): 20120155.CrossRefGoogle Scholar
  7. Connell, J. H., 1978. Diversity in tropical rain forests and coral reefs. Science 199: 1302–1310.CrossRefPubMedGoogle Scholar
  8. Damasceno-Junior, G. A., J. Semir, F. A. M. Dos Santos & H. de Freitas Leitão-Filho, 2005. Structure, distribution of species and inundation in a riparian forest of Rio Paraguai, Pantanal, Brazil. Flora-Morphology, Distribution, Functional Ecology of Plants 200: 119–135.CrossRefGoogle Scholar
  9. Davidson, T. A., A. W. Mackay, P. Wolski, R. Mazebedi, M. Murray-Hudson & M. Todd, 2012. Seasonal and spatial hydrological variability drives aquatic biodiversity in a flood-pulsed, sub-tropical wetland. Freshwater Biology 57: 1253–1265.CrossRefGoogle Scholar
  10. De Groot, R., L. Brander, S. Van Der Ploeg, R. Costanza, F. Bernard, L. Braat, M. Christie, N. Crossman, A. Ghermandi & L. Hein, 2012. Global estimates of the value of ecosystems and their services in monetary units. Ecosystem Services 1: 50–61.CrossRefGoogle Scholar
  11. Dubs, B., 1992. Observations on the differentiation of woodland and wet savanna habitats in the Pantanal of Mato Grosso, Brazil. Nature and dynamics of forest-savanna boundaries.-Chapman & Hall, London.Google Scholar
  12. European Commission, 2015. The EU Water Framework Directive – integrated river basin management for Europe – Environment – European Commission [WWW Document]. http://ec.europa.eu/environment/water/water-framework/index_en.html, Accessed on 3 Nov 15.
  13. Ferreira, L. V. & T. J. Stohlgren, 1999. Effects of river level fluctuation on plant species richness, diversity, and distribution in a floodplain forest in Central Amazonia. Oecologia 120: 582–587.CrossRefPubMedGoogle Scholar
  14. Foti, R., M. del Jesus, A. Rinaldo & I. Rodriguez-Iturbe, 2012. Hydroperiod regime controls the organization of plant species in wetlands. Proceedings of the National Academy of Sciences 109: 19596–19600.CrossRefGoogle Scholar
  15. Glenn, E. P., K. W. Flessa & J. Pitt, 2013. Restoration potential of the aquatic ecosystems of the Colorado River Delta, Mexico: introduction to special issue on “Wetlands of the Colorado River Delta”. Ecological Engineering 59: 1–6.CrossRefGoogle Scholar
  16. Grill, G., B. Lehner, A. E. Lumsdon, G. K. MacDonald, C. Zarfl & C. Reidy Liermann, 2015. An index-based framework for assessing patterns and trends in river fragmentation and flow regulation by global dams at multiple scales. Environmental Research Letters 10: 015001.CrossRefGoogle Scholar
  17. Gumbricht, T., J. McCarthy & T. S. McCarthy, 2004a. Channels, wetlands and islands in the Okavango Delta, Botswana, and their relation to hydrological and sedimentological processes. Earth Surface Processes and Landforms 29: 15–29.CrossRefGoogle Scholar
  18. Gumbricht, T., P. Wolski, P. Frost & T. S. McCarthy, 2004b. Forecasting the spatial extent of the annual flood in the Okavango delta, Botswana. Journal of Hydrology 290: 178–191.CrossRefGoogle Scholar
  19. Hamilton, S. K., 2010. Biogeochemical implications of climate change for tropical rivers and floodplains. Hydrobiologia 657: 19–35.CrossRefGoogle Scholar
  20. Hamilton, S. K., S. J. Sippel & J. M. Melack, 2002. Comparison of inundation patterns among major South American floodplains. Journal of Geophysical Research: Atmospheres 107: 8038.CrossRefGoogle Scholar
  21. Holtgrieve, G. W., M. E. Arias, K. N. Irvine, E. J. Ward, M. Kummu, J. Koponen, J. E. Richey & D. Lamberts, 2013. Ecosystem metabolism and support of freshwater capture fisheries in the Tonle Sap Lake, Cambodia. PLoS One 8: e71395.CrossRefPubMedPubMedCentralGoogle Scholar
  22. Jardine, T. D., N. R. Bond, M. A. Burford, M. J. Kennard, D. P. Ward, P. Bayliss, P. M. Davies, M. M. Douglas, S. K. Hamilton, J. M. Melack, R. J. Naiman, N. E. Pettit, B. J. Pusey, D. M. Warfe & S. E. Bunn, 2015. Does flood rhythm drive ecosystem responses in tropical riverscapes? Ecology 96: 684–692.CrossRefPubMedGoogle Scholar
  23. Junk, W. J. & M. T. F. Piedade, 1997. Plant life in the floodplain with special reference to herbaceous plants. The Central Amazon Floodplain. Ecology of a Pulsing System. Springer, Berlin: 147–186.CrossRefGoogle Scholar
  24. Junk, W.J. & Wantzen, K.M. 2004. The Flood pulse concept: new aspects, approaches, and applications – an update. In: Proceedings of the Second International Symposium on the Management of Large Rivers for Fisheries, RAP Publication 2004/17. FAO, Bangkok: 117–140.Google Scholar
  25. Junk, W.J., P.B. Bayley & Sparks, R.E., 1989. The flood pulse concept in river-floodplain systems. In: International Large River Symposium, Canadian Special Publication of Fisheries and Aquatic Sciences: 110–127.Google Scholar
  26. Junk, W. J., M. T. F. Piedade, J. Schöngart, M. Cohn-Haft, J. M. Adeney & F. Wittmann, 2011. A classification of major naturally-occurring Amazonian lowland wetlands. Wetlands 31: 623–640.CrossRefGoogle Scholar
  27. Junk, W. J., S. An, C. M. Finlayson, B. Gopal, J. Květ, S. A. Mitchell, W. J. Mitsch & R. D. Robarts, 2013. Current state of knowledge regarding the world’s wetlands and their future under global climate change: a synthesis. Aquatic Sciences 75: 151–167.CrossRefGoogle Scholar
  28. Kern, J. & Darwich, A., 1997. Nitrogen turnover in the Varzea. In: The central Amazon floodplain. Ecology of a pulsing system. Springer, Berlin: 119–136.Google Scholar
  29. Lehner, B. & P. Döll, 2004. Development and validation of a global database of lakes, reservoirs and wetlands. Journal of Hydrology 296: 1–22.CrossRefGoogle Scholar
  30. Lehner, B., C. R. Liermann, C. Revenga, C. Vörösmarty, B. Fekete, P. Crouzet, P. Döll, M. Endejan, K. Frenken, J. Magome, C. Nilsson, J. C. Robertson, R. Rödel, N. Sindorf & D. Wisser, 2011. High-resolution mapping of the world’s reservoirs and dams for sustainable river-flow management. Frontiers in Ecology and the Environment 9: 494–502.CrossRefGoogle Scholar
  31. Lesack, L. F. & J. M. Melack, 1995. Flooding hydrology and mixture dynamics of lake water derived from multiple sources in an Amazon floodplain lake. Water Resources Research 31: 329–345.CrossRefGoogle Scholar
  32. Lewis Jr, W. M., S. K. Hamilton, M. A. Lasi, M. Rodríguez & J. E. SAUNDERS III, 2000. Ecological determinism on the Orinoco floodplain. BioScience 50: 681–692.CrossRefGoogle Scholar
  33. Lobón-Cerviá, J., L. L. Hess, J. M. Melack & C. A. R. M. Araujo-Lima, 2015. The importance of forest cover for fish richness and abundance on the Amazon floodplain. Hydrobiologia 750: 245–255.CrossRefGoogle Scholar
  34. Mackey, R. L. & D. J. Currie, 2001. The diversity-disturbance relationship: is it generally strong and peaked? Ecology 82: 3479–3492.Google Scholar
  35. Malhi, Y., J. T. Roberts, R. A. Betts, T. J. Killeen, W. Li & C. A. Nobre, 2008. Climate change, deforestation, and the fate of the Amazon. Science 319: 169–172.CrossRefPubMedGoogle Scholar
  36. McDonald, J.A., P., Bunnat, P. Virak & Bunton, L., 1997. Plant communities of the Tonle Sap floodplain. UNESCO/IUCN/WI.Google Scholar
  37. Melack, J.M., & Hess, L.L., 2011. Remote sensing of the distribution and extent of wetlands in the Amazon basin. In: Amazonian floodplain forests. Springer, Berlin: 43–59.Google Scholar
  38. Mitsch, W. J. & J. W. Day Jr, 2006. Restoration of wetlands in the Mississippi–Ohio–Missouri (MOM) River Basin: experience and needed research. Ecological Engineering 26: 55–69.CrossRefGoogle Scholar
  39. Molino, J.-F. & D. Sabatier, 2001. Tree diversity in tropical rain forests: a validation of the intermediate disturbance hypothesis. Science 294: 1702–1704.CrossRefPubMedGoogle Scholar
  40. Murray-Hudson, M., 2009. Floodplain vegetation responses to flood regime in the seasonal Okavango Delta, Botswana (PhD thesis). University of Florida, Environmental Engineering Sciences, Gainesville, FL.Google Scholar
  41. Murray-Hudson, M., F. Combs, P. Wolski & M. T. Brown, 2011. A vegetation-based hierarchical classification for seasonally pulsed floodplains in the Okavango Delta, Botswana. African Journal of Aquatic Science 36: 223–234.CrossRefGoogle Scholar
  42. Murray-Hudson, M., P. Wolski, L. Cassidy, M. T. Brown, K. Thito, K. Kashe & E. Mosimanyana, 2015. Remote sensing-derived hydroperiod as a predictor of floodplain vegetation composition. Wetlands Ecology and Management 23: 603–616.CrossRefGoogle Scholar
  43. Naiman, R. J. & H. Décamps, 1997. The ecology of interfaces: riparian zones. Annual Review of Ecology and Systematics 28: 621–658.CrossRefGoogle Scholar
  44. Nesbitt, H. J., 1997. Rice production in Cambodia. International Rice Research Institute, Manila.Google Scholar
  45. Nilsson, C., C. A. Reidy, M. Dynesius & C. Revenga, 2005. Fragmentation and flow regulation of the world’s large river systems. Science 308: 405–408.CrossRefPubMedGoogle Scholar
  46. Nunes da Cunha, C. & W. J. Junk, 2001. Distribution of woody plant communities along the flood gradient in the Pantanal of Poconé, Mato Grosso, Brazil. International Journal of Ecology and Environmental Sciences 27: 63–70.Google Scholar
  47. Parolin, P., & Wittmann, F., 2010. Struggle in the flood: tree responses to flooding stress in four tropical floodplain systems. AoB Plants.Google Scholar
  48. Parolin, P., C. Lucas, M. T. F. Piedade & F. Wittmann, 2010. Drought responses of flood-tolerant trees in Amazonian floodplains. Annals of botany 105: 129–139.CrossRefPubMedGoogle Scholar
  49. Petraitis, P.S., R.E., Latham, & Niesenbaum, R.A., 1989. The maintenance of species diversity by disturbance. Quarterly Review of Biology 64: 393–418.CrossRefGoogle Scholar
  50. Petry, P., P. B. Bayley & D. F. Markle, 2003. Relationships between fish assemblages, macrophytes and environmental gradients in the Amazon River floodplain. Journal of Fish Biology 63: 547–579.CrossRefGoogle Scholar
  51. Pollock, M. M., R. J. Naiman & T. A. Hanley, 1998. Plant species richness in riparian wetlands—a test of biodiversity theory. Ecology 79: 94–105.Google Scholar
  52. Porter, J.W., & Muzila, I.L., 1989. Aspects of swamp hydrology in the Okavango. Botswana Notes and Records 21: 73–91.Google Scholar
  53. Pott, A., & Pott, V.J., 1994. Plantas do pantanal. Centro de Pesquisa Agropecuária do Pantanal, Serviço de Produção de Informação.Google Scholar
  54. Prance, G. T. & G. B. Schaller, 1982. Preliminary study of some vegetation types of the Pantanal, Mato Grosso, Brazil. Brittonia 34: 228–251.CrossRefGoogle Scholar
  55. Ratter, J. A., A. Pott, V. J. Pott, C. da Cunha & M. Haridasan, 1988. Observations on woody vegetation types in the Pantanal and at Corumbá, Brazil. Notes RBG, Edinburgh. 45 pp.Google Scholar
  56. Roxburgh, S. H., K. Shea & J. B. Wilson, 2004. The intermediate disturbance hypothesis: patch dynamics and mechanisms for species coexistence. Ecology 85: 359–371.CrossRefGoogle Scholar
  57. Rudorff, C. M., J. M. Melack & P. D. Bates, 2014. Flooding dynamics on the lower Amazon floodplain: 1. Hydraulic controls on water elevation, inundation extent, and river-floodplain discharge. Water Resources Research 50: 619–634.CrossRefGoogle Scholar
  58. Saatchi, S., W. Buermann, H. ter Steege, S. Mori & T. B. Smith, 2008. Modeling distribution of Amazonian tree species and diversity using remote sensing measurements. Remote Sensing of Environment 112: 2000–2017.CrossRefGoogle Scholar
  59. Sabo, J. L., R. Sponseller, M. Dixon, K. Gade, T. Harms, J. Heffernan, A. Jani, G. Katz, C. Soykan & J. Watts, 2005. Riparian zones increase regional species richness by harboring different, not more, species. Ecology 86: 56–62.CrossRefGoogle Scholar
  60. Sarkkula, J., J., Koponen, S., Hellsten, M., Keskinen, & Kiirikki, M., 2003. MRCS/WUP-FIN Model Report: Modelling Tonle Sap watershed and lake processes for environmental change assessment. MRC/WUP-FIN.Google Scholar
  61. Smith, P.A., 1976. An outline of the vegetation of the Okavango drainage system. Presented at the Symposium on the Okavango Delta and its future utilization, Botswana Society, National Museum, Gaborone.Google Scholar
  62. Snowy Mountains Engineering Corporation, 1989. Ecological Zoning Okavango Delta. Ministry of Local Government and Lands, Gaborone.Google Scholar
  63. Sullivan, P., E. Gaiser, D. Surratt, D. Rudnick, S. Davis & F. Sklar, 2014. Wetland ecosystem response to hydrologic restoration and management: the everglades and its urban-agricultural boundary (FL, USA). Wetlands 34: 1–8.CrossRefGoogle Scholar
  64. Tanentzap, A.J., W.G., Lee, & Schulz, K.A.C., 2013. Niches drive peaked and positive relationships between diversity and disturbance in natural ecosystems. Ecosphere 4: 133.Google Scholar
  65. Thorp, J. H., M. C. Thoms & M. D. Delong, 2006. The riverine ecosystem synthesis: biocomplexity in river networks across space and time. River Research and Applications 22: 123–147.CrossRefGoogle Scholar
  66. Van Trung, N., J.-H., Choi, & Won, J.-S., 2012. A land cover variation model of water level for the floodplain of Tonle Sap, Cambodia, derived from ALOS PALSAR and MODIS data. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 1–16.Google Scholar
  67. Warfe, D. M., N. E. Pettit, P. M. Davies, B. J. Pusey, S. K. Hamilton, M. J. Kennard, S. A. Townsend, P. Bayliss, D. P. Ward, M. M. Douglas, M. A. Burford, M. Finn, S. E. Bunn & I. A. Halliday, 2011. The “wet–dry” in the wet–dry tropics drives river ecosystem structure and processes in northern Australia. Freshwater Biology 56: 2169–2195.CrossRefGoogle Scholar
  68. Wassmann, R. & C. Martius, 1997. Methane emissions from the Amazon floodplain. The central Amazon floodplain. Ecology of a pulsing system. Springer, Berlin: 137–145.CrossRefGoogle Scholar
  69. Weber, G. E., 1997. Modelling nutrient fluxes in floodplain lakes. The central Amazon floodplain. Ecology of a pulsing system. Springer, Berlin: 109–117.CrossRefGoogle Scholar
  70. Wilson, B., & Dinçer, T., 1976. An introduction to the hydrology and hydrography of the Okavango Delta. Presented at the Symposium on the Okavango Delta and its future utilization, Botswana Society, National Museum, Gaborone.Google Scholar
  71. Winemiller, K. O., P. B. McIntyre, L. Castello, E. Fluet-Chouinard, T. Giarrizzo, S. Nam, I. G. Baird, W. Darwall, N. K. Lujan, I. Harrison, M. L. J. Stiassny, R. A. M. Silvano, D. B. Fitzgerald, F. M. Pelicice, A. A. Agostinho, L. C. Gomes, J. S. Albert, E. Baran, M. Petrere, C. Zarfl, M. Mulligan, J. P. Sullivan, C. C. Arantes, L. M. Sousa, A. A. Koning, D. J. Hoeinghaus, M. Sabaj, J. G. Lundberg, J. Armbruster, M. L. Thieme, P. Petry, J. Zuanon, G. T. Vilara, J. Snoeks, C. Ou, W. Rainboth, C. S. Pavanelli, A. Akama, A. van Soesbergen & L. Sáenz, 2016. Balancing hydropower and biodiversity in the Amazon, Congo, and Mekong. Science 351: 128–129.CrossRefPubMedGoogle Scholar
  72. Wittmann, F., 2012. Tree species composition and diversity in Brazilian freshwater floodplains. Nova Science Publishers, New York.Google Scholar
  73. Wittmann, F., D. Anhuf & W. J. Junk, 2002. Tree species distribution and community structure of central Amazonian várzea forests by remote-sensing techniques. Journal of Tropical Ecology 18: 805–820.CrossRefGoogle Scholar
  74. Wittmann, F., W. J. Junk & M. T. F. Piedade, 2004. The varzea forests in Amazonia: flooding and the highly dynamic geomorphology interact with natural forest succession. Forest Ecology and Management 196: 199–212.CrossRefGoogle Scholar
  75. Wittmann, F., J. Schongart, J. C. Montero, T. Motzer, W. J. Junk, M. T. F. Piedade, H. L. Queiroz & M. Worbes, 2006. Tree species composition and diversity gradients in white-water forests across the Amazon Basin. Journal of Biogeography 33: 1334–1347.CrossRefGoogle Scholar
  76. Wittmann, F., B. T. Zorzi, F. A. T. Tizianel, M. V. S. Urquiza, R. R. Faria, N. M. e Sousa, É. de Souza Módena, R. M. Gamarra & A. L. M. Rosa, 2008. Tree species composition, structure, and aboveground wood biomass of a riparian forest of the Lower Miranda River, southern Pantanal, Brazil. Folia Geobotânica 43: 397–411.CrossRefGoogle Scholar
  77. Wittmann, F., J. Schöngart & W. Junk, 2011. Phytogeography, species diversity, community structure and dynamics of central Amazonian floodplain forests. In Junk, W. J., M. T. F. Piedade, F. Wittmann, J. Schöngart & P. Parolin (eds), Amazonian floodplain forests, ecological studies. Springer, Dordrecht: 61–102.Google Scholar
  78. Wittmann, F., E. Householder, M. T. Piedade, R. L. de Assis, J. Schöngart, P. Parolin & W. J. Junk, 2013. Habitat specifity, endemism and the neotropical distribution of Amazonian white-water floodplain trees. Ecography 36: 690–707.CrossRefGoogle Scholar
  79. Zeilhofer, P. & M. Schessl, 2000. Relationship between vegetation and environmental conditions in the northern Pantanal of Mato Grosso, Brazil. Journal of Biogeography 27: 159–168.CrossRefGoogle Scholar
  80. Zelnik, I. & A. Čarni, 2008. Distribution of plant communities, ecological strategy types and diversity along a moisture gradient. Community Ecology 9: 1–9.CrossRefGoogle Scholar
  81. Zurbrügg, R., J. Wamulume, R. Kamanga, B. Wehrli & D. B. Senn, 2012. River-floodplain exchange and its effects on the fluvial oxygen regime in a large tropical river system (Kafue Flats, Zambia). Journal of Geophysical Research: Biogeosciences 117: G03008.CrossRefGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2016

Authors and Affiliations

  1. 1.Sustainability Science ProgramHarvard UniversityCambridgeUSA
  2. 2.Department of BiogeochemistryMax Planck Institute for ChemistryMainzGermany
  3. 3.University of HamburgHamburgGermany
  4. 4.Okavango Research InstituteUniversity of BotswanaMaunBotswana
  5. 5.Department of Civil and Natural Resources EngineeringUniversity of CanterburyChristchurchNew Zealand

Personalised recommendations