Next Article in Journal
Effects of Five Planting Cover Measures on Soil Crust Particle Size Distribution Characteristics in Ulan Buh Desert
Next Article in Special Issue
Wetland Biodiversity and Ecosystem Conservation: Integrating Genetic, Species, and Ecosystem Perspectives for Effective Action
Previous Article in Journal
A Dataset of Pipunculidae and Psilidae (Diptera) in Some Regions of European Russia
Previous Article in Special Issue
Do Ecological Patterns Persist in Highly Impacted Urban Wetlands? A Spatiotemporal Analysis of Aquatic Macrophytes and Limnological Variability in a Peruvian Coastal Wetland
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Diversity
Volume 18
Issue 5
10.3390/d18050274
diversity-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Patterns in Understorey Vegetation of a Semi-Arid Terminal Wetland over 20 Years in Response to Flood and Drought

by
Rebekah Grieger
1,*,
Jaiden Johnston-Bates
1,
Andres Sutton
2 and
Samantha J. Capon
3
1
Australian Rivers Institute, Griffith University, Nathan, QLD 4111, Australia
2
Centre for Ecosystem Science (CES), School of Biological Earth and Environmental Science (BEES), University of New South Wales, Sydney, NSW 2052, Australia
3
Griffith Institute for Human and Environmental Resilience, School of Environment and Science, Griffith University, Nathan, QLD 4111, Australia
*
Author to whom correspondence should be addressed.
Diversity 2026, 18(5), 274; https://doi.org/10.3390/d18050274
Submission received: 28 February 2026 / Revised: 30 April 2026 / Accepted: 30 April 2026 / Published: 1 May 2026
(This article belongs to the Special Issue Wetland Biodiversity and Ecosystem Conservation)
Browse Figures
Versions Notes

Abstract

Floodplains are key components of inland river systems of Australia with floodplain vegetation playing important roles in habitat provision, nutrient cycling, and supporting strong cultural values. These vegetation communities are highly dynamic, particularly in response to flooding. However, decades of water development and highly managed water resources are linked to wetland habitat decline in this region. We explored patterns of vegetation response to flooding over twenty years at the Narran Lakes Ramsar site, a terminal floodplain wetland system in the northern Murray–Darling Basin, Australia. We collated data from previous monitoring efforts and resampled permanent plots for understorey vegetation structure and composition. Three flood events were surveyed over a 20-year period, with each event surveyed on two occasions first, following initial drawdown (minimal standing water) and a second survey under dry or drier conditions (~6 months after the recession of floodwaters). Overall, we observed a high diversity of native plant species (~110 species) in understorey communities across the wetland and high compositional turnover both between flood events and within years (i.e., paired surveys). Notably, vegetation cover, but not species richness, was greatest in the 2023 survey following the largest of the three flood events investigated. Understorey composition was strongly driven by inundation regimes, particularly the duration of recent inundation, and the number of wet and dry years prior. Large flood events are critical for supporting vegetation resilience in these systems, increasingly so under a drier climate and with stretched water resources. Continued long-term monitoring of vegetation through flood cycles at the Narran Lakes will be critical to understanding ecological responses to longer-term changes in climate and hydrology to inform adaptive water management and maintain the values of this Ramsar site.

1. Introduction

Inland floodplain wetlands are highly dynamic ecosystems, both spatially and temporally, largely driven by hydrology. Periods of flooding punctuate otherwise dry conditions, which range from regular periods of dry to atypical periods of severe drought [1]. Temporal and spatial variation in inland floodplain wetland vegetation occurs due to variations in flow [2,3]. Changes in the frequency, duration, magnitude, timing, regularity and rate of change of flow events influences species composition [4,5]. Floodplain wetland vegetation is adapted to such variable conditions and many taxa exhibit traits which enable their persistence through dry periods or otherwise unfavourable conditions (e.g., persistent seed banks and physical dormancy [6,7]. Natural flow regimes have been widely altered in large river systems globally for anthropogenic uses such as for agriculture, mining and for drinking water [8,9]. The capacity of individuals, species, and communities to tolerate drought or flooding is critically important where altered hydrological regimes expose plants to highly variable and uncharacteristic moisture conditions [10]. Floodplain wetland vegetation is typically considered resilient to these naturally large perturbations in hydrology. However, the limits to their resilience under altered flow regimes and climatic extremes remain unknown.
The capacity of vegetation communities to recover following natural or anthropogenically driven disturbances is a key component of wetland resilience [11]. A general definition of ecological resilience is the ability of a system to maintain or restore community structure and function when faced with disturbance [12]. Inland floodplain wetland vegetation communities tend to exhibit both wet and dry phases in terms of species representation and productivity [13], but can rapidly reorder from a dry to a wet phase after a flood event [14] which contributes to their overall resilience. Each consecutive phase may not share the same mix of species but rather shows similarities in total plant cover and the expression of functional groups (e.g., annual forbs, perennial shrubs) [5,15]. This dynamism in vegetation composition and productivity suggests that an extreme external disturbance would be required to transition the system into an alternate state. Th ability of inland floodplain wetlands to switch between phases is underpinned by life-history strategies of plant species, including the formation of persistent soil seed banks, phenological flexibility, and traits that enable establishment under the pressure of, or following, disturbance [4,16].
Reordering of vegetation between wet and dry phases and over time can be measured through temporal species turnover, i.e., the species which are gained or lost over time [17,18]. Alternating wet and dry cycles in semi-arid wetlands drive species turnover and replacement, particularly among annual herbaceous species that respond rapidly to water and resource fluctuations [19]. This turnover is dictated by the filtering of species, where environmental conditions exclude those lacking traits suited to the prevailing conditions [20]. For example, flood pulses trigger germination and growth of a range of moisture-dependent taxa [21], while intervening dry periods favour terrestrial, stress-tolerant species capable of persisting under conditions of low moisture availability [22,23,24]. Stochastic recruitment events following an influx of moisture availability further contribute to temporal variability in community composition [25]. Across temporal scales, cyclical, seasonal and variable periods of hydrology drive short-term shifts in community structure which subsequently accumulate to produce broader long-term trends in the composition and structure of vegetation communities [26]. Long-term vegetation trajectories in terminal wetlands therefore reflect the cumulative impact of repeated wet–dry transitions [27,28] and contribute to the overall resilience of such systems in terms of their capacity to respond to changing conditions and to maintain functionality, e.g., biomass production.
The development of water resources and climate change pose significant disruption to hydrological processes that sustain inland floodplain wetlands [8,29,30]. Upstream, the extraction, regulation and alteration of natural flow regimes have already driven changes to the frequency, duration, and magnitude of overbank flooding across many semi-arid systems [8,29,30]. In the northern Murray–Darling Basin, reduced surface water availability and reduced connection between river and floodplain systems has occurred as a result of damming, extraction, and the reduction in small-to-medium flow events [31]. These changes have contributed to the dieback of tree populations [32], reductions in the extent of non-woody wetlands, and changes to the composition of soil seed banks [23,33]. Climate change can compound impacts of flow alteration by intensifying hydrological extremes, lengthening drought periods, and increasing evapotranspiration [34,35]. Future climate modelling suggests reductions in flow up to 8% with up to a 40% decrease in rainfall runoff in regions of the northern basin by 2030 [18,35,36]. Managing available water resources to achieve environmental and economic outcomes will be vital under increasingly water-scarce conditions.
Understanding patterns of vegetation dynamics and wetland condition over time is important for managing wetlands and water resources into the future. Monitoring of riparian and wetland conditions has been growing in Australia, particularly since 2012 under government-funded programmes such as Flow-MER (Flow-Monitoring Evaluation and Research program) and LTIM (Long-Term Intervention Monitoring) which have been established to assess the outcomes of environmental watering actions for target biota [37,38]. However, much focus has been on systems in the southern Murray–Darling Basin where streamflow is greater, more consistent, and water inputs are highly managed through dam releases and regulation infrastructure [39]. In contrast, river systems within the northern basin have more variable flow regimes, fewer large storages, and less regulation [39]. Several monitoring projects within the northern basin have been conducted over shorter timescales (i.e., 2–3 years up to 10 years) [40]; however, understanding of key ecological processes, such as response to altered flow regimes, restoration, or fire, can occur over much longer timeframes [41]. In floodplain wetlands, vegetation responds to hydrological regimes which can vary in the timing, duration, frequency, and rate of change, such that systems can be derived from the unique combination of flow regimes over history. Limited work exists exploring vegetation responses to repeated flood events through time [37] and this lack of long-term data is identified as a key limiting factor to adequate water management [15,42,43]. Satellite imagery can provide insight to landscape scale dynamics through the use of vegetation and water indices [15,42,43] such as the normalised difference vegetation index (NDVI), enhanced vegetation index (EVI), and normalised difference water index (NDWI), but these are not able to provide information regarding the compositional characteristics of vegetation communities.
Here, we explore patterns in understorey vegetation structure and composition in response to three flood events, interspersed by drought, across a 20-year period in the semi-arid wetland complex of the Narran Lakes. More specifically, we examine patterns in understorey vegetation structure and composition between and within flood event survey periods (i.e., paired surveys following drawdown and following six months of subsequent drying) to determine the role of flow in short- and long-term vegetation community dynamics for this semi-arid wetland. We also explore dynamics of floodplain wetland vegetation communities over this 20-year period through the lens of local and landscape-scale species turnover and the role of flood events in compositional change. Based on the existing knowledge of inland floodplain wetland vegetation dynamics, we expected to see high plant cover and greater species richness, particularly for annual species, in drawdown surveys and a reduction in these metrics following drying. Temporal changes are less well known; however, we expected that flood events resulting in similar hydrological conditions would result in similar vegetation responses.

2. Materials and Methods

2.1. Study Area

The Narran Lakes wetland complex (herein Narran), Dharriwaa in the language of the Yuwaalarayy/Euahlayi First Nations people, is a terminal floodplain system of the Narran River in the northern Murray–Darling Basin (Figure 1; ref. [44]). The wetland complex, covering an area of at least 40,900 ha, comprises several lakes (Clear Lake, Back Lake, North Arm, Narran Lake), and areas of vegetated wetlands. The main Narran River channel flows through the northern wetland section, diverting into the northern lakes before spilling into Narran Lake proper and the southern wetlands and floodplains [44,45]. Only in rare, very large floods does the Narran River connect through the wetland to the Barwon and Bokhara Rivers (south west). More commonly, the northern lakes dry out in approximately 4–8 months through evaporation and underground seepage [46]. A portion of the northern wetlands (8447 ha) within the Narran Lake Nature Reserve is listed as a Ramsar wetland.
The climate is semi-arid with low annual rainfall (440 mm long-term annual average), hot summers (average maximum temperatures between 21 and 36 °C) and mild winters (3.6–18 °C [47]). Rainfall is highly variable; however, widespread flooding of the wetland is typically driven by monsoonal summer rainfall events in the upper Condamine–Balonne catchment [48]. Historically, the northern lakes would fill every 1.5–2.5 years, flood events were long and slow to recede, while river flow was highly variable and frequently experienced cease-to-flow and dry conditions. Upstream water resource development (i.e., dams, diversions, irrigation) has resulted in reduced mean annual flow and reduced flood duration and frequency, especially for smaller and moderate inflow events, which are critical for the wetlands [45]. Recent (2000–2024) mean annual flow for the Narran River is approximately 90,000 ML, although it is highly variable and annual flows of up to 370,000 ML have been recorded and periods of extended nil flows have also been recorded [49].

2.2. Vegetation Communities

Three broad vegetation classes dominate the wetland ecosystems of Narran Lakes [50]: floodplain woodlands, lignum shrublands and non-woody wetlands. Floodplain woodlands are typically dominated by Eucalyptus camaldulensis (river red gum), E. coolabah (coolabah), a sub-canopy of Acacia stenophylla (river coobah) and Eremophila bignoniiflora (emu bush), and an understorey of mixed grasses, forbs, and shrubs. Lignum shrublands are dominated by varying densities of the large, clump-forming shrub Duma florulenta (tangled lignum), and a variable understorey of inundation-responding forbs, sedges, and grasses with a sparse canopy of A. stenophylla. Lignum shrublands are key habitats for waterbirds which nest on the large woody structure of lignum during floods. Non-woody wetlands typically occur along lake margins or lakebeds when water recedes and are dominated by mixed forbs, sedges and grasses, typically annual species which respond rapidly to the drawdown conditions. This study focuses on vegetation dynamics within lignum shrublands and non-woody wetlands as the responses of understorey communities are comparable in these areas which lack a substantial tree canopy.

2.3. Survey Design

Field data were collected across six survey periods between 2004 and 2024 which represented paired surveys across three flood events: first after initial drawdown of floodwaters and a second survey after six months of drying (Table 1, Figure 2). As the focus of this study we selected ten sites across the wetland which were surveyed in all flood periods of 2004–2005 (flood 1 [45]), 2008–2009 (flood 2), and 2023–2024 (flood 3, see [51]). Sites represented two broad vegetation types, a variety of flood frequencies (i.e., 4–17 years flooded of 20 years), and three geographical regions (non-woody wetland: 3 sites; lignum shrubland: 7 sites; Figure 1). Additional lignum shrubland and non-woody wetland sites monitored more broadly within the wetland complex, although not included in this analysis, also fall within this range of flood frequency. We have excluded sites in tree-dominated vegetation classes due to the effect of canopy cover on understorey responses and the lack of long-term monitoring sites [51].

2.4. Antecedent Environmental Conditions

Across the 20-year study period, hydro-climatic conditions were highly variable capturing a period of significant drought (2001–2009) and extensive flooding (2020–2024). Conditions preceding the 2023–2024 surveys were significantly wetter and for a longer sustained period than the conditions preceding the earlier surveys which followed small inundation events (Table 1, Figure 2, Supplementary Figure S1). Surveys in November 2004 and 2008, and August 2023, represented the period after drawdown of Clear Lake when flows had ceased for approximately 180 days. However, rainfall prior to the 2008 survey supplemented water levels in Clear Lake while rainfall was low preceding August 2023 (Table 1). Surveys in May 2005, April 2009, and February 2024 represented the period after 6 months of drying, although rainfall prior to May 2005 was lower. In February 2024, no inflows were recorded for close to one year, however recent local and upstream rainfall provided increased flows into Clear Lake during the survey (Table 1). In November 2023, between the surveys for flood 3, a wildfire burned three shrubland sites on the western floodplain, all of which experienced low-to-moderate burn severities according to national mapping [46].

2.5. Field Data Collection

During each vegetation survey we assessed understorey vegetation communities within a 20 × 20 m (0.04 ha) plot at each site. We assessed understorey (i.e., less than 1 m height) vegetation cover in 10 randomly distributed 1 × 1 m quadrats within the larger 0.04 ha site, recording the percentage cover for each species.
Each species observed was identified to the finest taxonomic level possible (genus or species) in the field or using images and samples following the nomenclature of the Australian Plant Census and descriptions of PlantNet [53] and Cunningham et al. [54]. Species were later categorised by plant functional group (i.e., tree, shrub, forb, grass, sedge, fern), life history (i.e., annual, perennial), and endemicity (i.e., native, non-native). Species names were checked against the Australian Plant Census to update older observations with current nomenclature and ensure consistency across the surveys.

2.6. Environmental Data

Spatial data for surface water inundation in target areas (i.e., a 30 m radius from each vegetation survey site) was sourced from LandSat 5, 7, 8 and 9 satellites through the Digital Earth Australia sandbox. Images were first filtered to remove timesteps with less than 90% coverage of the target area and cloud cover. Relative area of inundation was determined for each timestep using the Fisher Water Index [55] as the percentage of the area with values greater than −10.63. This threshold was defined using Receiver Operating Characteristic curves [55,56] and reference inundation maps of Narran Lakes floodplain between 2022 and 2024 [57,58]. For each water year (i.e., July to June), we then calculated the start day of inundation events (day of year), and the duration of the inundation event at each site. These metrics were then summarised across the study period (2004–2024) to capture long-term flood frequency and duration as well as short-term effects (Table 2).

2.7. Data Analysis

To describe the floristic composition of understorey wetland vegetation of NLNR and differences over time, we summarised observed species by family, genera, plant growth form, life history and endemicity as total species richness and average vegetation cover. We also identified the number of surveys and floods (i.e., paired surveys) each species and genera was observed in, the number of species unique in each survey, and the number of new species to each survey. For each site we then we calculated species richness as the number of different species observed within the 10 quadrats. We also calculated the site’s average vegetation cover for all species, growth forms, and native and exotic species.
We assessed temporal species turnover using the approach of Baselga and Orme [59] with the Sorensen family of indices, whereby beta diversity (i.e., the ratio of gamma and alpha diversity) is decomposed to turnover (i.e., the Simpson dissimilarity, species replacement) and nestedness (i.e., the nestedness of Sorensen, a subset of the original species pool). We first presence-/absence-transformed the species observation matrix then calculated all site × survey pairwise comparisons using the function beta.multi from the betapart R package (version 1.6.1) [60]. We then filtered the resulting dataset to only include temporal comparisons by site (i.e., excluding comparisons of site A to site B at any time) and plotted the mean of comparisons across all surveys and floods.
To explore patterns in species composition within and between survey periods, habitat types and in relation to flood history variables, we used redundancy analysis (RDA). We first Hellinger-transformed species cover data and added a dummy species (x = 1) to all rows to account for zero-inflated species composition data. We standardised all numeric explanatory variables to zero-mean and unit variance using decostand (vegan R package version 2.7-1 [61]) with the standardise method and checked for collinearity using variance inflation factor (VIF), discarding variables with VIF > 10 (Supplementary Table S2). We also included the factorial variable Survey Date and Season to capture temporal and seasonal variation and binary variables for flooding in the previous year and burned in the November 2023 fire (Table 2). We used forward stepwise model selection to reduce the number of model terms. Model and variable significance was tested with analysis of variance (anova.cca; vegan), following the protocol of Borcard et al. [62].
The results of RDA were extracted and plotted with ggplot2, where sites were plotted as points, numerical environmental variables were plotted as vectors and factorial environmental variables were plotted as points. Using scaling 2, the length and direction of environmental vectors reflect their importance to the ordination axes and each other (i.e., vectors in opposite directions are negatively correlated [62]).
To further describe variation in species composition we used the multivariate homogeneity of group dispersion method [63]. We calculated beta-dispersion (betadisper; vegan R package) on a Simpson dissimilarity matrix of species composition for all site × survey combinations, then tested for differences is dispersion between sites using analysis of variance.

3. Results

3.1. General Floristics

At least 110 taxa (87 genera) were observed in the understorey across the ten study sites across all surveys, representing 40 families, a large proportion of which were native (~78%) or exhibited a forb growth form (~72%). Native species of the family Chenopodiaceae were highly represented (12 species), followed by 11 native and four exotic species of the family Asteraceae, seven native Poaceae species, six species of Malvaceae (three native, three exotic), and five native Cyperaceae species. Exotic species were present in the families Asteraceae (four species (sp.)), Malvaceae (three sp.), Verbenaceae (three sp.), Boraginaceae (two sp.), and one species each of Carophyllaceae, Convulvulaceae, Fabaceae, Primulaceae, and Rosaceae.
The annual forb (38 native sp., 14 exotic) and perennial forb (23 native sp., four exotic) plant functional groups were the dominant plant forms across all surveys. Annual and perennial monocots (i.e., grasses and sedges) were represented by five and eight species respectively. Shrubs were represented by nine species and two tree species (Eucalyptus camaldulensis and Acacia stenophylla) were present as seedlings in the groundcover strata. Commonly observed exotic species included Cuscuta campestris (six surveys), Conyza bonariensis (five surveys), while Abutilon theophrastii and Verbena macrostachya occurred in four surveys. The prevalence of these exotic species within the studied sites has changed over time, most notably for C. bonariensis which was only recorded in one site in Nov–04 but increased to five sites during Aug–23 and Feb–24 (Supplementary Table S1). In contrast, V. macrostachya was observed in six sites during Nov–04 which reduced to three sites by Feb–24.

3.2. Changes in Diversity in Response to Flood Events/Pulses

Only six species were recorded from these sites in all surveys (Supplementary Table S1) including two perennial forbs (Calotis scapigera and Marsilea drummondii), the exotic annual forb Cascuta campestris, perennial monocot Sporobolus mitchellii, and shrub Duma florulenta. Thirty species, including five exotic species, were observed across all floods (i.e., at least one of the paired flood surveys), half of which were annual forbs (four exotic), followed by seven perennial forbs (one exotic), three perennial monocots, three shrubs, and two annual monocots. Twenty-four species (three exotic) were observed in two of the three floods, while 56 species (10 exotic) were only observed in one flood, 80% of which were only observed in one survey. The greatest number of exotic species were observed during November 2004 (26 sp., three unique) followed by August 2023 (21 sp., one unique; Table 3).
No species were exclusively observed in multiple dry surveys (i.e., May–05, Apr–09, Feb–24), although 10 species were observed in a single dry survey, seven of which were recorded in Feb–24. In contrast, several species were observed exclusively in drawdown surveys (i.e., Nov–04, Nov–08, Aug–23) including nine annual forbs and two annual monocots (Lachnagrostis filiformis, Isolepis australis). New species were observed during all surveys, with greater numbers of new and unique species (i.e., only observed in one survey) drawdown surveys (Figure 3A,B). Over time, species richness varied, particularly between the flood pairs with richness greatest in drawdown surveys. Overall, a marginal decline in the number of species observed is apparent over time when considering the paired flood events (Figure 3C,D).

3.3. Patterns in Vegetation Structure over Time

Total vegetation cover, functional group cover, and native or exotic taxa cover varied substantially over time (Figure 4). In general, total vegetation cover was greatest during drawdown surveys and dominated by annual forbs. Total cover in dry surveys was the lowest in May–05, intermediate in April–09 and high in Feb–24, higher even than total cover in the earlier drawdown surveys (Figure 4A). Perennial sub-shrub and perennial shrub cover was greatest in Aug–23 and Feb–24 compared to previous surveys. Similarly, perennial monocot cover also showed greater cover in Feb–24, while the cover of annual forbs, perennial forbs, and annual monocots was reduced from Aug–23 (Figure 4A). Cover of native and exotic taxa also varied across surveys, with greater cover of native and exotic taxa during drawdown surveys compared to dry surveys (Figure 4B). Exotic taxa showed greater cover during the Aug–23 survey but was similar to that observed during earlier surveys again by Feb–24.

3.4. Change in Composition over Time

Species turnover of understorey taxa within sites, as represented by binary Simpson dissimilarity, generally increased over time, but varied between sites (Figure 5). Western floodplain sites (WS_1, WS_2) exhibited the greatest increase in turnover, particularly with substantial differences in surveys 5 and 6 compared to all others. Turnover at the sites NWW_1, NWW_2, and WS_3 followed a similar pattern and showed a substantial difference particularly in relation to survey 2 (May–05). The sites ES_1 and ES_4 exhibited moderate turnover for all surveys. Northern wetland sites (i.e., ES_2, ES_3 and NWW_3) also increased in turnover over time, peaking in survey 5 (Aug–23, Figure 5). The pattern of increasing turnover with time is also present when comparing flood events (i.e., species turnover between floods for composition combined by survey pairs) where composition during the first and second floods are similar while the composition during flood 3 shows greater difference to both earlier events (Figure 5). The alternate is true for WS_3 where composition during flood 3 is equal to both earlier events while flood 2 is different from flood 1.

3.5. Drivers of Compositional Change

Redundancy analysis identified a significant relationship between forward selected historical flooding metrics and understorey species composition which explained approximately 21% (adjusted R2) of variation in composition (F = 2.73, p = 0.001, Figure 6, Supplementary Table S2). In general, axis one (RDA1) represented a gradient of short-term flood history associated with the duration of inundation (F = 2.25, p = 0.004) and greater long-term flood frequency (F = 8.58, p = 0.001) on the right-hand side (Figure 6A). Axis two (RDA2) partially represented a gradient of short-term flood history and survey dates, extending from surveys with greater consecutive years of inundation (F = 1.69, p = 0.032) on the positive axis direction to less frequent inundation in the negative (Figure 6A). Survey date was also a significant contributor to the model (F = 2.07, p = 0.001) highlighting the compositional variation between individual surveys (Figure 6A) and flood events (Figure 6B). Composition within sites also varied over time (Figure 6B), tending to be more homogenous in 23–24 surveys compared to 04–05 surveys and 08–09 surveys which appeared to form two distinct community types of eastern wetland sites (non-woody wetland (NWW) and eastern shrubland (ES)) on the right-hand side, and western shrubland (WS) sites on the left-hand side (Figure 6B) during the earlier surveys. Fire in western shrubland sites also contributed to species composition (F = 1.9, p = 0.013).
Dispersion (i.e., the ordination distance between time points) of each site across survey dates (i.e., within site variation) was relatively similar for all sites, suggesting equal and large variation in species composition between surveys regardless of site (Supplementary Table S3).

4. Discussion

Wetland vegetation communities surveyed across the Narran Lakes wetland system have varied substantially in structure and composition across the 20-year study period in response to wetting and drying. Vegetation in lignum shrublands and non-woody wetlands has changed over time both within individual flood periods, as drying occurs, and between successive flood events, at both site and whole-wetland scales. Most notably, species richness and plant cover at a site scale, and the number of new species observed overall, were greater in drawdown surveys than during subsequent dry surveys, highlighting the importance of flood events for replenishing plant biodiversity [64]. High levels of species turnover between survey periods and the discovery of new species during each survey highlights the episodic and dynamic nature of these vegetation communities, with over half of the taxa recorded overall only observed during a single flood event. Annual forbs varied most between surveys, particularly with respect to their cover abundance, reflecting their highly responsive nature to environmental conditions. Understorey species composition was strongly associated with flood history, both short- and long-term, especially the duration of recent flooding, the number of wet years prior, and the frequency of flooding over the study period. Survey date and the recent fire also contributed to species composition, further highlighting the temporal variability of this system. These results together highlight the spatially and temporally dynamic nature of this inland floodplain wetland system and the vital role of flood events (both small and large) in wetland dynamics [14].
Much of the existing work on wetland vegetation in the Murray–Darling Basin has focused on short-term (i.e., up to 5 years) responses to individual flooding events [37]. In contrast, our 20-year dataset captures multiple discrete flood cycles, enabling us to examine not only immediate post-flood responses but also vegetation dynamics over longer timescales, facilitating a comparison of responses across multiple flood events. Across individual events, our findings support expected patterns reported in previous studies (e.g., [6,7,33,40]), such as substantial increases in species richness and plant cover immediately following floodwater drawdown with subsequent declines as a result of drying. Taxa with annual life histories showed the greatest fluctuation in richness and abundance between drawdown and dry surveys. Annuals which maintain a persistent soil seed bank can lay dormant for long dry periods and respond rapidly to flood drawdown, often proliferating under these high-soil-moisture conditions and rapidly completing their life cycles [7,65]. Expression of the bulk of soil seed bank diversity often occurs after floods and typically comprises a highly diverse and abundant assemblage of annual taxa [6]. The dominance of annuals observed in drawdown surveys here suggests that soil seed banks strongly contribute to the diversity of extant assemblages immediately following floodwater drawdown while the perennial taxa become more dominant as the wetland dries out [5,6,23].
A key finding of this study was the exceptionally high species turnover observed across both space and time. New taxa were recorded in every survey, with many species detected only once or restricted to a single flood event (i.e., paired surveys). This raises questions regarding the origin of these newly recorded species and whether they represent recent establishment events or the emergence of individuals from persistent local sources (i.e., soil seed and propagule banks). Two main sources are likely: in situ sources (i.e., soil seed banks, asexual reproduction, dispersal within the wetland complex) and ex situ sources (i.e., dispersal from upstream, transport by animals or wind [66]). Soil seed banks are recognised as important seed stores in dynamic systems, particularly for those which experience hydrological variability like the Narran Lakes and wetlands within the Murray–Darling Basin, where they are often distinct from the extant vegetation but largely comprise annual taxa [6,16,33], which we observed in the drawdown surveys.
Dispersal of taxa within the wetland complex is also likely, with many of the newly observed species noted more widely throughout the studied vegetation types and woodland vegetation types within the wetland, including exotic species [45,51,67]. The large flood event in 2022 likely dispersed propagules throughout the Narran wetland complex, from core, frequently flooded areas to fringing habitats and vice versa, while also providing conditions suitable for germination and establishment of a greater suite of perennials than was achieved in the earlier, smaller floods. Alternatively, ex situ propagule sources such as dispersal from upstream is likely for taxa which are hydrochorous [68,69] or those dispersed by animals which visit the wetland, i.e., waterbirds who nest in the lignum during large flood events [70]. Dispersal and establishment of exotic taxa is commonly noted following flood disturbances [68] and accordingly we recorded greater richness and cover of exotics during drawdown surveys. However, the number of new exotics was similar for all floods and reduced in dry surveys, suggesting the exotics were not able to survive under drying conditions. Similar numbers of exotic plant species have been recorded in other, more northerly MDB Ramsar-listed wetlands [40]; however, their management would require a catchment-wide approach. Further assessment of the soil seed bank assemblage and dispersal patterns is required to understand their roles in species turnover and vegetation dynamics.
Other factors influencing the vegetation responses observed here could include local canopy or shrub layer conditions [71], rainfall, temperature, rate of drying, grazing [65], or other local disturbances (i.e., fire). Temperature and rainfall can alter germination cues and influence the composition of emerging taxa [72], which could have contributed to the different responses between dry surveys where rainfall was higher in the preceding months. Additionally, the cooler winter temperatures experienced during the August 2023 drawdown survey may have resulted in different species being observed, associated with seasonal temperature differences (i.e., a greater number of winter annuals) [72,73] and low rainfall. However, given our study design incorporates flood events, we cannot detangle seasonal effects from hydrology and further study is required to understand vegetation responses following smaller flood events and under different temperature scenarios—although Wen et al. [42] suggests that hydrology would have a greater influence than season on these wetlands given their inherit dynamism and flood-responsive nature. Local canopy conditions (i.e., canopy cover) and the presence of shrub cover can also influence the abundance and diversity of understorey taxa [71,74]. While we did not directly measure these attributes and explored sites without a dominant tree canopy, the record of increased perennial shrub, sub-shrub, and tree taxa within understorey quadrats recorded during the third flood surveys suggests greater cover in other strata which may have influenced the expression of understorey vegetation by providing shade and reducing evaporation, leading to rapid senescence of plant material [74]. Localised disturbances, such as a fire which occurred in November 2023, could also have contributed to the high turnover between the Aug–23 and Feb–24 surveys observed in western shrubland sites by reducing standing plant biomass, removing the germinable soil seed bank and litter seed bank, or triggering the gemination of fire-responding species [75,76]. However, burned sites were compositionally similar to those unburned by Feb–24, suggesting that rainfall supported recovery. Understanding how this event has altered vegetation communities and the role of flooding for restoration requires further investigation; however, Mackay et al. [76], suggest that fire impacts may only be temporary, particularly if followed by inundation, and that inundation history is a stronger driver of composition and structure compared to short-term fire effects.
Floodplain wetland vegetation communities are structured by, and respond to, short- and long-term patterns in flood history. We observed highly variable but native-dominated vegetation communities through space and time which allow the system to persist through abiotic perturbations (i.e., water fluctuations and disturbance events) and respond with a suite of species adapted to specific conditions. Notably, plant cover, but not richness, was greatest during the latest flood event, likely as a result of multiple wet years (Figure 2) supporting the growth of specific species. The smaller flood pulses which occurred prior to vegetation surveys in 2004–05 and 2008–09 maintained heterogenous communities across the eastern and western floodplains. The larger flood event surveyed in 2023–24, both in terms of inundation extent and duration, resulted in large turnover across most sites and a more homogenous community assemblage at the whole-wetland scale. These results highlight the importance of variable flood events for maintaining a mosaic of diverse wetland vegetation communities [77,78,79], further supporting long-term resilience. Smaller events provide an important resource pulse to the core wetland area which can drive divergence in species composition between the core area and those on the periphery. Larger events can reset wetland systems [24] by facilitating the distribution and germination of taxa across all areas of the wetland, resulting in a more homogenous assemblage at an ecosystem scale. Subsequent drying or smaller flood events which only reach core areas would likely result again in diverging communities across the wetland system [24]. This variability in the magnitude, duration, and frequency of flood events drives the distribution and resilience of vegetation communities in arid floodplain wetlands at local and landscape scales [79]. Heterogeneity within and between vegetation communities at the landscape scale [78] and over time contributes to their resilience by providing a large species and resource pool from which local extirpations can be replenished, particularly during flood events.
Floodplain wetland plant communities are inherently dynamic, switching between wet and dry states [13,14], which contributes to their overall resilience. However, this process is challenged under extreme drying, altered flow regimes, and climate change. If a drier climate develops in the Narran Lakes region, as expected under many climate change scenarios [35,80], significant ecological impacts are likely to follow, including dieback of canopy trees, altered plant community composition through local extirpations or invasions, and the degradation of vegetation condition [80]. Significant drying could also facilitate encroachment by terrestrial vegetation [78,81] or an increase in the cover of woody species as we have observed. If drying is coupled with anthropogenic pressures of land modification and water extraction, there may be a reduced capacity to return such terrestrialised areas to a wetland state without management intervention. Despite experiencing a significant drought event in 2013–2019 (Figure 2), vegetation communities studied here maintained high native richness following the flood with minimal change in the richness and cover of exotics species, further highlighting their ability to recover when water returns. Flow management, through prescribed environmental flows that mimic more natural flow patterns, can support biodiversity and ecological processes [82]. The strategic implementation of environmental flow regimes can be used to simulate the natural variability of both short and long inundation events, both of which we have demonstrated are pivotal to the resilience of floodplain wetland vegetation communities [14,21]. Specifically, such flooding events nourish extant vegetation by meeting moisture requirements, stimulate germination from seed banks and facilitate seed dispersal via hydrochory, and provide connectivity between disjointed habitats [7,16,69]. Managed environmental flows are mechanisms proven to achieve beneficial outcomes for wetland environments in many regions of the Murray–Darling Basin, including the Narran Lakes [67]. Despite challenges in environmental water delivery for this region, such as highly variable flows, small managed water storages, large private storages, and flow-contingent environmental water licences [83], the use of environmental flows in conjunction with natural flow events is highly valuable. We strongly recommend the continued use of environmental watering actions, both through flow protection and managed releases from public and private storages, in this system. These actions, and monitoring of their outcomes, are particularly important in the face of a drying climate; however, we highlight the need for variability in managed flow regimes to maintain vegetation mosaics [84].
Understanding of wetland vegetation patterns and responses to flow events over long periods is lacking, particularly for the northern Murray–Darling Basin where observational evidence is rarely reported over more than 5 years [37]. While there is a growing body of work, much of the reporting tends to identify changes across a single flood event or responses to a drought breaking [85]. Long-term monitoring is particularly important for understanding changes to vegetation structure and dynamics of woody taxa [41,81], which in turn influences the understorey community, although we have not included tree-dominated sites in our analysis. Here, we instead present evidence across multiple flood events and suggest that ‘vital sign’ monitoring [40] in conjunction with reactionary monitoring provides the most holistic view of vegetation resilience and responses. We also acknowledge that funding schemes and personnel changes can hamper efforts of long-term monitoring; there is much-needed value and certainty in management decisions that can result from this understanding [86]. Although our results align with other studies of understorey vegetation in semi-arid wetlands [6,7,33,40], caution in interpretation is required given the small number of sites analysed and the small proportion of variance in species composition captured by the redundancy model. Additionally, the results presented are the culmination of multiple disturbance events (i.e., floods, droughts, fire) and rather than attempting to draw out the individual effects we instead suggest that this presents a realistic view of wetland vegetation responses in variable systems, which are likely to become more variable under future climate projections [39].
This 20-year study demonstrates that wetland vegetation at the Narran Lakes is inherently dynamic, structured by the magnitude, duration, and sequencing of flood events. Large inundation events acted as system-wide reset mechanisms, reshaping composition and redistributing species across the landscape. Exceptionally high species turnover and continual recruitment from seed banks and dispersal pathways underscore the pulse-driven nature of these communities. Through these many fluctuations the lignum shrublands and non-woody wetlands of the Narran Lakes system have demonstrated a high level of dynamism by maintaining substantial native biodiversity and functional composition, despite challenges of altered flow regimes, drought, and fire. Maintaining variability in flood regimes, rather than simply restoring water, is therefore fundamental to sustaining biodiversity, ecosystem function, and vegetation responses in arid wetlands under a changing climate.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/d18050274/s1, Figure S1: Rainfall and river flow history for January 2000–February 2024. Table S1: Summary of species observed across the period 2004—2024 at selected long term monitoring sites within the Narran Lake Nature Reserve. Table S2: Environmental variables used to model variation in species composition through redundancy analysis showing variance inflation factor, constrained variance and term significance. Table S3: Average dispersion of sites over time calculated from Simpson dissimilarity of species cover for understorey vegetation.

Author Contributions

Conceptualization, R.G. and S.J.C.; Methodology, R.G., A.S. and S.J.C.; Formal analysis, R.G.; Data curation, R.G., J.J.-B., A.S. and S.J.C.; Writing—original draft, R.G. and J.J.-B.; Writing—review & editing, all authors; Funding acquisition, S.J.C. All authors have read and agreed to the published version of the manuscript.

Funding

This work was funded by the Australian Government Department of Climate Change Energy Environment and Water through the Commonwealth Environmental Water Holder. Earlier work contributing data to this paper was funded through the Murray Darling Basin Authority and the Commonwealth Environmental Water Holder.

Data Availability Statement

Data is contained within the article or Supplementary Material. Further inquiries can be directed to the corresponding authors.

Acknowledgments

We acknowledge the traditional owners of the land on which this study was conducted, the Yuwaalarayy/Euahlayi peoples, who have a deep and continuing cultural connection with the land and waters. We thank the many volunteers who contributed to field surveys and we acknowledge New South Wales Parks and Wildlife and the Narran Lakes Joint Management Committee for logistical support and cultural knowledge sharing. We also acknowledge Peta Zivec for review of an earlier version of this manuscript and the anonymous reviewers for their contributions.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Jiao, T.; Williams, C.A.; Rogan, J.; De Kauwe, M.G.; Medlyn, B.E. Drought Impacts on Australian Vegetation During the Millennium Drought Measured with Multisource Spaceborne Remote Sensing. J. Geophys. Res. Biogeosciences 2020, 125, e2019JG005145. [Google Scholar] [CrossRef]
  2. Ballinger, A.; Nally, R.M. The Landscape Context of Flooding in the Murray–Darling Basin. In Advances in Ecological Research; Academic Press: Cambridge, MA, USA, 2006; Volume 39, pp. 85–105. [Google Scholar]
  3. Sheldon, F.; Boulton, A.J.; Puckridge, J.T. Conservation value of variable connectivity: Aquatic invertebrate assemblages of channel and floodplain habitats of a central Australian arid-zone river, Cooper Creek. Biol. Conserv. 2002, 103, 13–31. [Google Scholar] [CrossRef]
  4. Casanova, M.T.; Brock, M.A. How do depth, duration and frequency of flooding influence the establishment of wetland plant communities? Plant Ecol. 2000, 147, 237–250. [Google Scholar] [CrossRef]
  5. Wassens, S.; Ning, N.; Hardwick, L.; Bino, G.; Maguire, J. Long-term changes in freshwater aquatic plant communities following extreme drought. Hydrobiologia 2017, 799, 233–247. [Google Scholar] [CrossRef]
  6. Capon, S.J.; Reid, M.A. Vegetation resilience to mega-drought along a typical floodplain gradient of the southern Murray-Darling Basin, Australia. J. Veg. Sci. 2016, 27, 926–937. [Google Scholar] [CrossRef]
  7. James, C.S.; Capon, S.J.; White, M.G.; Rayburg, S.C.; Thoms, M.C. Spatial variability of the soil seed bank in a heterogeneous ephemeral wetland system in semi-arid Australia. Plant Ecol. 2007, 190, 205–217. [Google Scholar] [CrossRef]
  8. Nilsson, C.; Reidy, C.A.; Dynesius, M.; Revenga, C. Fragmentation and Flow Regulation of the World’s Large River Systems. Science 2005, 308, 405–408. [Google Scholar] [CrossRef]
  9. Grill, G.; Lehner, B.; Thieme, M.; Geenen, B.; Tickner, D.; Antonelli, F.; Babu, S.; Borrelli, P.; Cheng, L.; Crochetiere, H.; et al. Mapping the world’s free-flowing rivers. Nature 2019, 569, 215–221, Erratum in Nature 2019, 572, E9.. [Google Scholar] [CrossRef]
  10. Niinemets, Ü. Responses of forest trees to single and multiple environmental stresses from seedlings to mature plants: Past stress history, stress interactions, tolerance and acclimation. For. Ecol. Manag. 2010, 260, 1623–1639. [Google Scholar] [CrossRef]
  11. Middleton, B.A. Soil seed banks and the potential restoration of forested wetlands after farming. J. Appl. Ecol. 2003, 40, 1025–1034. [Google Scholar] [CrossRef]
  12. Holling, C.S. Resilience and Stability of Ecological Systems. Annu. Rev. Ecol. Evol. Syst. 1973, 4, 1–23. [Google Scholar] [CrossRef]
  13. Thapa, R.; Thoms, M.C.; Parsons, M.; Reid, M. Adaptive cycles of floodplain vegetation response to flooding and drying. Earth Surf. Dynam. 2016, 4, 175–191. [Google Scholar] [CrossRef]
  14. Bunn, S.E.; Thoms, M.C.; Hamilton, S.K.; Capon, S.J. Flow variability in dryland rivers: Boom, bust and the bits in between. River Res. Appl. 2006, 22, 179–186. [Google Scholar] [CrossRef]
  15. Thapa, R.; Thoms, M.C.; Parsons, M. The response of dryland floodplain vegetation productivity to flooding and drying. J. Arid Environ. 2016, 129, 42–55. [Google Scholar] [CrossRef]
  16. Kehr, J.M.; Merritt, D.M.; Stromberg, J.C. Linkages between primary seed dispersal, hydrochory and flood timing in a semi-arid region river. J. Veg. Sci. 2014, 25, 287–300. [Google Scholar] [CrossRef]
  17. Dornelas, M.; Magurran, A.E.; Buckland, S.T.; Chao, A.; Chazdon, R.L.; Colwell, R.K.; Curtis, T.; Gaston, K.J.; Gotelli, N.J.; Kosnik, M.A.; et al. Quantifying temporal change in biodiversity: Challenges and opportunities. Proc. Biol. Sci. 2013, 280, 20121931. [Google Scholar] [CrossRef] [PubMed]
  18. Shimadzu, H.; Dornelas, M.; Magurran, A.E. Measuring temporal turnover in ecological communities. Methods Ecol. Evol. 2015, 6, 1384–1394. [Google Scholar] [CrossRef]
  19. Stott, I.; Franco, M.; Carslake, D.; Townley, S.; Hodgson, D. Boom or bust? A comparative analysis of transient population dynamics in plants. J. Ecol. 2010, 98, 302–311. [Google Scholar] [CrossRef]
  20. Portela, A.P.; Durance, I.; Vieira, C.; Honrado, J. Environmental filtering and environmental stress shape regional patterns of riparian community assembly and functional diversity. Freshw. Biol. 2023, 68, 1428–1441. [Google Scholar] [CrossRef]
  21. Brock, M.; Casanova, M. Plant life at the edge of wetlands: Ecological responses to wetting and drying patterns. Front. Ecol. Res. Build. Links 1997, 181–192. [Google Scholar] [CrossRef]
  22. Bried, J.T.; Jog, S.K.; Dzialowski, A.R.; Davis, C.A. Potential Vegetation Criteria for Identifying Reference-Quality Wetlands in the South-Central United States. Wetlands 2014, 34, 1159–1169. [Google Scholar] [CrossRef]
  23. Nicol, J.M.; Ganf, G.G.; Pelton, G.A. Seed banks of a southern Australian wetland: The influence of water regime on the final floristic composition. Plant Ecol. 2003, 168, 191–205. [Google Scholar] [CrossRef]
  24. Reynolds, L.V.; Cooper, D.J.; Hobbs, N.T. Drivers of riparian tree invasion on a desert stream. River Res. Appl. 2014, 30, 60–70. [Google Scholar] [CrossRef]
  25. Gui, H.; Hou, L.; Wang, J.; Dong, X.; Han, S. Flood changed the community composition and increased the importance of stochastic process of vegetation and seed bank in a riparian ecosystem of the Yellow River. Ecol. Indic. 2023, 154, 110505. [Google Scholar] [CrossRef]
  26. Poff, N.L.; Allan, J.D.; Bain, M.B.; Karr, J.R.; Prestegaard, K.L.; Richter, B.D.; Sparks, R.E.; Stromberg, J.C. The Natural Flow Regime. BioScience 1997, 47, 769–784. [Google Scholar] [CrossRef]
  27. Capon, S.J. Plant community responses to wetting and drying in a large arid floodplain. River Res. Appl. 2003, 19, 509–520. [Google Scholar] [CrossRef]
  28. Busch, D.E.; Loftus, W.F.; Bass, O.L. Long-term hydrologic effects on marsh plant community structure in the southern Everglades. Wetlands 1998, 18, 230–241. [Google Scholar] [CrossRef]
  29. Sheldon, F.; Rocheta, E.; Steinfeld, C.; Colloff, M.J.; Moggridge, B.; Carmody, E.; Hillman, T.; Kingsford, R.T.; Pittock, J. Are environmental water requirements being met in the Murray–Darling Basin, Australia? Mar. Freshw. Res. 2024, 75, MF23172. [Google Scholar] [CrossRef]
  30. Davies, B.R.; Thoms, M.C.; Walker, K.F.; O’Keeffe, J.H.; Gore, J.A. Dryland Rivers: Their Ecology, Conservation and Management. In The Rivers Handbook; Blackwell Scientific: Oxford, UK, 1994; pp. 484–511. [Google Scholar]
  31. Kingsford, R.T. Ecological impacts of dams, water diversions and river management on floodplain wetlands in Australia. Austral Ecol. 2000, 25, 109–127. [Google Scholar] [CrossRef]
  32. Catelotti, K.; Kingsford, R.T.; Bino, G.; Bacon, P. Inundation requirements for persistence and recovery of river red gums (Eucalyptus camaldulensis) in semi-arid Australia. Biol. Conserv. 2015, 184, 346–356. [Google Scholar] [CrossRef]
  33. Reid, M.; Capon, S. Role of the soil seed bank in vegetation responses to environmental flows on a drought-affected floodplain. River Syst. 2011, 19, 249–259. [Google Scholar] [CrossRef]
  34. Döll, P.; Zhang, J. Impact of climate change on freshwater ecosystems: A global-scale analysis of ecologically relevant river flow alterations. Hydrol. Earth Syst. Sci. 2010, 14, 783–799. [Google Scholar] [CrossRef]
  35. Murray-Darling Basin Authority. The Future of Climate and Water Availability in the Murray-Darling Basin; Murray-Darling Basin Authority: Canberra, Australia, 2025. [Google Scholar]
  36. CSIRO. Water Availability in the Condamine-Balonne: A Report to the Australian Government from the CSIRO Murray-Darling Basin Sustainable Yeilds Project; CSIRO: Canberra, Australia, 2008. [Google Scholar]
  37. Colloff, M.J.; Caley, P.; Saintilan, N.; Pollino, C.A.; Crossman, N.D. Long-term ecological trends of flow-dependent ecosystems in a major regulated river basin. Mar. Freshw. Res. 2015, 66, 957–969. [Google Scholar] [CrossRef]
  38. Gawne, B.; Hale, J.; Stewardson, M.J.; Webb, J.A.; Ryder, D.S.; Brooks, S.S.; Campbell, C.J.; Capon, S.J.; Everingham, P.; Grace, M.R.; et al. Monitoring of environmental flow outcomes in a large river basin: The Commonwealth Environmental Water Holder’s long-term intervention in the Murray–Darling Basin, Australia. River Res. Appl. 2020, 36, 630–644. [Google Scholar] [CrossRef]
  39. Chiew, F.; Zheng, H.; Post, D.; Robertson, D.; Rojas, R. Hydroclimate trends and future projections in the Murray-Darling Basin. In Technical Report for the Murray-Darling Basin Authority; CSIRO: Canberra, Australia, 2022. [Google Scholar]
  40. Mason, T.J.; Honeysett, J.; Thomas, R.F.; Popovic, G.C.; Hosking, T.; Shelly, D.J.; Bowen, S. Monitoring vital signs: Wetland vegetation responses to hydrological resources in the Macquarie Marshes NSW, Australia. Austral Ecol. 2022, 47, 1296–1314. [Google Scholar] [CrossRef]
  41. Lindenmayer, D.B.; Lavery, T.; Scheele, B.C. Why We Need to Invest in Large-Scale, Long-Term Monitoring Programs in Landscape Ecology and Conservation Biology. Curr. Landsc. Ecol. Rep. 2022, 7, 137–146. [Google Scholar] [CrossRef]
  42. Wen, L.; Mason, T.J.; Ryan, S.; Ling, J.E.; Saintilan, N.; Rodriguez, J. Monitoring long-term vegetation condition dynamics in persistent semi-arid wetland communities using time series of Landsat data. Sci. Total Environ. 2023, 905, 167212. [Google Scholar] [CrossRef]
  43. Sims, N.C.; Colloff, M.J. Remote sensing of vegetation responses to flooding of a semi-arid floodplain: Implications for monitoring ecological effects of environmental flows. Ecol. Indic. 2012, 18, 387–391. [Google Scholar] [CrossRef]
  44. Butcher, R.; Hale, J.; Capon, S.J.; Thoms, M.C. Ecological character description for Narran Lake Nature Reserve Ramsar Site Ecological Character Description. In Report to the Department of Sustainability, Environment, Water, Population and Communities; Commonwealth of Australia: Canberra, Australia, 2011. [Google Scholar]
  45. Thoms, M.; Capon, S.; James, C.; Padgham, M.; Rayburg, S. The Narran Ecosystem Project: The Response of a Terminal Wetland System to Variable Wetting and Drying. Final Report to the Murray–Darling Basin Commission; Murray-Darling Basin Commission: Canberra, Australia, 2007. [Google Scholar]
  46. Thomas, R.; Karunaratne, S.; Heath, J.; Kuo, W. Assessment of Site-Specific Flow Indicator Inundated Areas in the Vegetation of Narran Lakes; NSW Office of Environment and Heritage: Parramatta, Australia, 2016. [Google Scholar]
  47. BOM. Climate Data Online, Daily Rainfall for Lightning Ridge (48243). Available online: https://www.bom.gov.au/climate/data/?ref=ftr (accessed on 28 March 2024).
  48. Rayburg, S.; Thoms, M. A coupled hydraulic–hydrologic modelling approach to deriving a water balance model for a complex floodplain wetland system. Hydrol. Res. 2009, 40, 364–379. [Google Scholar] [CrossRef]
  49. BOM. Water Data Online, Narran River at Wilby Wilby (Belvedere) 244016. Available online: https://www.bom.gov.au/waterdata/ (accessed on 6 June 2024).
  50. McGann, T.D.; Kingswood, R.; Bell, D. Vegetation of Narran Lake Nature Reserve, North Western Plains, New South Wales. Cunninghamia 2001, 7, 43–63. [Google Scholar]
  51. Grieger, R.; Johnston-Bates, J.; Capon, S. Long-Term Wetland Vegetation Dynamics of Dharriwaa-Narran Lakes Nature Reserve: In Report for the Commonwealth Environmental Water Holder; Department of Climate Change, Energy, the Environment and Water: Brisbane, Australia, 2024. [Google Scholar]
  52. Dunn, B.; Ai, E.; Alger, M.J.; Fanson, B.; Fickas, K.C.; Krause, C.E.; Lymburner, L.; Nanson, R.; Papas, P.; Ronan, M.; et al. Wetlands Insight Tool: Characterising the Surface Water and Vegetation Cover Dynamics of Individual Wetlands Using Multidecadal Landsat Satellite Data. Wetlands 2023, 43, 37. [Google Scholar] [CrossRef]
  53. Royal Botanic Gardens and Domain Trust. PlantNet (The NSW Plant Information Network System). Available online: https://plantnet.rbgsyd.nsw.gov.au (accessed on 6 June 2024).
  54. Cunningham, G.M.; Mulham, W.E.; Milthorpe, P.L.; Leigh, J.H. Plants of Western New South Wales; CSIRO Publishing: Sydney, Australia, 2011. [Google Scholar]
  55. Fisher, A.; Flood, N.; Danaher, T. Comparing Landsat water index methods for automated water classification in eastern Australia. Remote Sens. Environ. 2016, 175, 167–182. [Google Scholar] [CrossRef]
  56. Fawcett, T. An introduction to ROC analysis. Pattern Recognit. Lett. 2006, 27, 861–874. [Google Scholar] [CrossRef]
  57. Heath, J.T.; Grimmett, L.; Gopalakrishnan, T.; Thomas, R.F.; Lenehan, J. Integrating Sentinel 2 Imagery with High-Resolution Elevation Data for Automated Inundation Monitoring in Vegetated Floodplain Wetlands. Remote Sens. 2024, 16, 2434. [Google Scholar] [CrossRef]
  58. State Government of NSW; NSW DCCEEW. Inundation Maps for NSW Inland Floodplain Wetlands; New South Wales Department of Climate Change Energy the Environment and Water: Sydney, Australia, 2020. [Google Scholar]
  59. Baselga, A.; Orme, C.D.L. betapart: An R package for the study of beta diversity. Methods Ecol. Evol. 2012, 3, 808–812. [Google Scholar] [CrossRef]
  60. Baselga, A. Partitioning the turnover and nestedness components of beta diversity. Glob. Ecol. Biogeogr. 2010, 19, 134–143. [Google Scholar] [CrossRef]
  61. Oksanen, J.; Simpson, G.L.; Blanchet, F.G.; Kindt, R.; Legendre, P.; Minchin, P.R.; O’Hara, R.B.; Solymos, P.; Stevens, M.H.H.; Szoecs, E.; et al. vegan: Community Ecology Package. CRAN R Package, 2025. Available online: https://cran.r-project.org/web/packages/vegan/index.html (accessed on 29 April 2026). [CrossRef]
  62. Borcard, D.; Gillet, F.; Legendre, P. Spatial Analysis of Ecological Data. In Numerical Ecology with R; Borcard, D., Gillet, F., Legendre, P., Eds.; Springer International Publishing: Cham, Switzerland, 2018; pp. 299–367. [Google Scholar]
  63. Anderson, M.J.; Ellingsen, K.E.; McArdle, B.H. Multivariate dispersion as a measure of beta diversity. Ecol. Lett. 2006, 9, 683–693. [Google Scholar] [CrossRef] [PubMed]
  64. Leigh, C.; Sheldon, F.; Kingsford, R.T.; Arthington, A.H. Sequential floods drive ‘booms’ and wetland persistence in dryland rivers: A synthesis. Mar. Freshw. Res. 2010, 61, 896–908. [Google Scholar] [CrossRef]
  65. Reid, M.A.; Ogden, R.; Thoms, M.C. The influence of flood frequency, geomorphic setting and grazing on plant communities and plant biomass on a large dryland floodplain. J. Arid Environ. 2011, 75, 815–826. [Google Scholar] [CrossRef]
  66. Zivec, P.; Johnston-Bates, J. Seed rain as a propagule source for restoration of semi-arid floodplain old fields. Appl. Veg. Sci. 2024, 27, e70001. [Google Scholar] [CrossRef]
  67. Vincent, B.; Bowen, S.; Southwell, M. Vegetation Community and Tree Stand Condition Assessment of the Flood-Dependent vegetation of Narran Lakes Nature Reserve, 2020–2021. In A Report to the Commonwealth Environmental Water Office; Department of Climate Change, Energy, the Environment and Water: Canberra, Australia, 2022. [Google Scholar]
  68. Catford, J.A.; Jansson, R. Drowned, buried and carried away: Effects of plant traits on the distribution of native and alien species in riparian ecosystems. New Phytol. 2014, 204, 19–36. [Google Scholar] [CrossRef]
  69. Hyslop, J.; Trowsdale, S. A review of hydrochory (seed dispersal by water) with implications for riparian rehabilitation. J. Hydrol. 2012, 51, 137–152. [Google Scholar]
  70. Brandis, K.; Bino, G.; Spencer, J.; Ramp, D.; Kingsford, R. Decline in colonial waterbird breeding highlights loss of Ramsar wetland function. Biol. Conserv. 2018, 225, 22–30. [Google Scholar] [CrossRef]
  71. Capon, S.J.; Balcombe, S.R.; McBroom, J. Environmental watering for vegetation diversity outcomes must account for local canopy conditions. Ecohydrology 2017, 10, e1859. [Google Scholar] [CrossRef]
  72. Ooi, M.K.J.; Auld, T.D.; Denham, A.J. Climate change and bet-hedging: Interactions between increased soil temperatures and seed bank persistence. Glob. Change Biol. 2009, 15, 2375–2386. [Google Scholar] [CrossRef]
  73. Capon, S.J.; Brock, M.A. Flooding, soil seed bank dynamics and vegetation resilience of a hydrologically variable desert floodplain. Freshw. Biol. 2006, 51, 206–223. [Google Scholar] [CrossRef]
  74. James, C.S.; Capon, S.J.; Quinn, G.P. Nurse plant effects of a dominant shrub (Duma florulenta) on understorey vegetation in a large, semi-arid wetland in relation to flood frequency and drying. J. Veg. Sci. 2015, 26, 985–994. [Google Scholar] [CrossRef]
  75. Bixby, R.J.; Cooper, S.D.; Gresswell, R.E.; Brown, L.E.; Dahm, C.N.; Dwire, K.A. Fire effects on aquatic ecosystems: An assessment of the current state of the science. Freshw. Sci. 2015, 34, 1340–1350. [Google Scholar] [CrossRef]
  76. Mackay, K.D.; Vincent, B.; Southwell, M.; Growns, I.; Mika, S. Differential impacts of fire and inundation on a wetland plant community after wildfire. Mar. Freshw. Res. 2024, 75, MF24040. [Google Scholar] [CrossRef]
  77. Lawson, J.R.; Fryirs, K.A.; Lenz, T.; Leishman, M.R. Heterogeneous flows foster heterogeneous assemblages: Relationships between functional diversity and hydrological heterogeneity in riparian plant communities. Freshw. Biol. 2015, 60, 2208–2225. [Google Scholar] [CrossRef]
  78. Shilpakar, R.L.; Thoms, M.C.; Reid, M.A. The resilience of a floodplain vegetation landscape. Landsc. Ecol. 2021, 36, 139–157. [Google Scholar] [CrossRef]
  79. Raulings, E.J.; Morris, K.A.Y.; Roache, M.C.; Boon, P.I. The importance of water regimes operating at small spatial scales for the diversity and structure of wetland vegetation. Freshw. Biol. 2010, 55, 701–715. [Google Scholar] [CrossRef]
  80. Johnston-Bates, J.; Grieger, R.; Capon, S.J. Predicting trajectories of dryland wetland vegetation transformation under climate change: A case study of the northern Murray–Darling Basin, Australia. Mar. Freshw. Res. 2024, 75, MF24016. [Google Scholar] [CrossRef]
  81. Saintilan, N.; Asbridge, E.; Lucas, R.; Rogers, K.; Wen, L.; Powell, M.; Colloff, M.J.; Rodriguez, J.F.; Saco, P.M.; Sandi, S.; et al. Australian forested wetlands under climate change: Collapse or proliferation? Mar. Freshw. Res. 2021, 73, 1255–1262. [Google Scholar] [CrossRef]
  82. Acreman, M.; Arthington, A.H.; Colloff, M.J.; Couch, C.; Crossman, N.D.; Dyer, F.; Overton, I.; Pollino, C.A.; Stewardson, M.J.; Young, W. Environmental flows for natural, hybrid, and novel riverine ecosystems in a changing world. Front. Ecol. Environ. 2014, 12, 466–473. [Google Scholar] [CrossRef]
  83. DPIE. Intersecting Streams Long Term Water Plan Parts A and B; Department of Planning, Industry and Environment: Parramatta, Australia, 2020. [Google Scholar]
  84. Campbell, C.J.; Freestone, F.L.; Duncan, R.P.; Higgisson, W.; Healy, S.J. The more the merrier: Using environmental flows to improve floodplain vegetation condition. Mar. Freshw. Res. 2021, 72, 1185–1195. [Google Scholar] [CrossRef]
  85. Kingsford, R.T.; Mac Nally, R.; King, A.; Walker, K.F.; Bino, G.; Thompson, R.; Wassens, S.; Humphries, P. A commentary on ‘Long-term ecological trends of flow-dependent ecosystems in a major regulated river basin’, by Matthew J. Colloff, Peter Caley, Neil Saintilan, Carmel A. Pollino and Neville D. Crossman. Mar. Freshw. Res. 2015, 66, 970–980. [Google Scholar] [CrossRef]
  86. Dourado, G.F.; Rallings, A.M.; Viers, J.H. Overcoming persistent challenges in putting environmental flow policy into practice: A systematic review and bibliometric analysis. Environ. Res. Lett. 2023, 18, 043002. [Google Scholar] [CrossRef]
Diversity 18 00274 g001
Figure 1. Map of Narran Lakes field survey sites in relation to the Narran Lakes wetland complex (main map), the Condamine–Balonne River catchment (top right), and the Murray–Darling Basin (top left). Site colour relates to the regions of ES: eastern shrubland, NWW: non-woody wetland, WS: western shrubland.
Figure 1. Map of Narran Lakes field survey sites in relation to the Narran Lakes wetland complex (main map), the Condamine–Balonne River catchment (top right), and the Murray–Darling Basin (top left). Site colour relates to the regions of ES: eastern shrubland, NWW: non-woody wetland, WS: western shrubland.
Diversity 18 00274 g001
Diversity 18 00274 g002
Figure 2. Fractional cover of water, vegetation, and soil for Narran Lakes Nature Reserve showing the timing of vegetation surveys after the drawdown of floodwaters as red triangles. Data generated in Digital Earth Australia sandbox using the Wetland Insights Tool (version 4.0) [52].
Figure 2. Fractional cover of water, vegetation, and soil for Narran Lakes Nature Reserve showing the timing of vegetation surveys after the drawdown of floodwaters as red triangles. Data generated in Digital Earth Australia sandbox using the Wetland Insights Tool (version 4.0) [52].
Diversity 18 00274 g002
Diversity 18 00274 g003
Figure 3. Metrics of understorey wetland taxa over time. (A,B) Total species richness across all sites by flood (A) and by survey (B), showing stacked values of new observations, new and unique observations (i.e., new in that survey and not in any other survey), and previously observed. (C,D) Average species richness per site over time by flood (C) and by survey (D). (E,F) Average number of native and exotic taxa per site over time by flood (E) and by survey (F). Error bars represent standard error from the mean.
Figure 3. Metrics of understorey wetland taxa over time. (A,B) Total species richness across all sites by flood (A) and by survey (B), showing stacked values of new observations, new and unique observations (i.e., new in that survey and not in any other survey), and previously observed. (C,D) Average species richness per site over time by flood (C) and by survey (D). (E,F) Average number of native and exotic taxa per site over time by flood (E) and by survey (F). Error bars represent standard error from the mean.
Diversity 18 00274 g003
Diversity 18 00274 g004
Figure 4. Stacked barplot of average vegetation cover per site. (A) Cover of plant functional groups over time. (B) Cover of native and exotic taxa over time. Total bar height represents total plant cover average by site. Cover values can be greater than 100 to account for taxa overlapping.
Figure 4. Stacked barplot of average vegetation cover per site. (A) Cover of plant functional groups over time. (B) Cover of native and exotic taxa over time. Total bar height represents total plant cover average by site. Cover values can be greater than 100 to account for taxa overlapping.
Diversity 18 00274 g004
Diversity 18 00274 g005
Figure 5. Turnover of presence-/absence-transformed species composition over time by site as represented by Simpson dissimilarity of pairwise comparisons to all previous surveys. (A) Turnover between all surveys, (B) turnover between floods (i.e., change between flood 1 (04–05) and flood 2 (08–09) shown for time 3). Shapes represent the survey order of the pairwise comparison to all previous surveys (i.e., circle represents the comparison of subsequent surveys to the first survey). Line represents the mean turnover value at each time point.
Figure 5. Turnover of presence-/absence-transformed species composition over time by site as represented by Simpson dissimilarity of pairwise comparisons to all previous surveys. (A) Turnover between all surveys, (B) turnover between floods (i.e., change between flood 1 (04–05) and flood 2 (08–09) shown for time 3). Shapes represent the survey order of the pairwise comparison to all previous surveys (i.e., circle represents the comparison of subsequent surveys to the first survey). Line represents the mean turnover value at each time point.
Diversity 18 00274 g005
Diversity 18 00274 g006
Figure 6. Redundancy analysis biplot for understorey species composition in relation to forward selected environmental variables; axis titles show the proportion of constrained variance explained. (A) All sites across surveys showing the forward selected environmental variables. (B) Median centroids of sites in paired flood surveys showing trajectory over time as the flood number ordered; all points shown in background. Flood frequency: long-term flood frequency; Duration: duration of inundation in previous year; consecutive flood: consecutive years inundation; Burned_N and _Y: burned in November 2023 no and yes. Asterisks represent significance level of environmental variable (* p < 0.05, *** p < 0.001). Numbers within points represent the flood order 1: Nov-04–May-05; 2: Nov-08–Apr-09; 3: Aug-23–Feb-24.
Figure 6. Redundancy analysis biplot for understorey species composition in relation to forward selected environmental variables; axis titles show the proportion of constrained variance explained. (A) All sites across surveys showing the forward selected environmental variables. (B) Median centroids of sites in paired flood surveys showing trajectory over time as the flood number ordered; all points shown in background. Flood frequency: long-term flood frequency; Duration: duration of inundation in previous year; consecutive flood: consecutive years inundation; Burned_N and _Y: burned in November 2023 no and yes. Asterisks represent significance level of environmental variable (* p < 0.05, *** p < 0.001). Numbers within points represent the flood order 1: Nov-04–May-05; 2: Nov-08–Apr-09; 3: Aug-23–Feb-24.
Diversity 18 00274 g006
Table 1. Antecedent hydrological conditions for six months preceding field surveys. Rainfall and temperature data for Lightning Ridge post office (station 48243), flow data for Wilby Wilby gauge on the Narran River (gauge 422016), drawdown from Clear Lake and inundation extent determined from LandSat Water Observations (Digital Earth Australia).
Table 1. Antecedent hydrological conditions for six months preceding field surveys. Rainfall and temperature data for Lightning Ridge post office (station 48243), flow data for Wilby Wilby gauge on the Narran River (gauge 422016), drawdown from Clear Lake and inundation extent determined from LandSat Water Observations (Digital Earth Australia).
SurveySeasonTemperature
Minimum–Maximum		Total Rainfall (mm)Mean Daily Flow (ML/d)Cumulative Flow (ML)Days After Cease to FlowMonths After Drying of Clear Lake Inundation Extent (ha)Classification
Flood 1
November 2004 Early summer (flood Feb–July 2004)9.6 (±5.3)–23.6 (±5.7)~16000~2001 (Oct 04)2624Drawdown
May 2005 Autumn18.8 (±4.3)–32.8 (±4.5)~260~20~430~3807 (Oct 04)99Dry
Flood 2
November 2008 Early summer (flood Jan–March 2008)9.4 (±6.2)–24.1 (±5.7)~2500~5~19001278Drawdown
April 2009 Autumn 18.5 (±4.5)–32.8 (±4.8)~370~110~1970~705 (Dec 08)1470Dry
Flood 3
August 2023Winter (flood 2022–Jan 2023)12.1 (±6.9)–25.9 (±6.7)~90~20~4290~1600770 (February 2023, maximum 4550 in 2022)Drawdown
February 2024 Late summer16.9 (±6.3)–32.2 (±5.6)~330~30~8700~320 (3 days of flow > 1000 ML as surveying commenced)5 (Sept 23) filling1703 (late Feb 2024)Dry
Table 2. Summary of environmental variables and flood history metrics calculated for each site during all surveys from field data and the Fisher Water Index applied to LandSat satellite images during the study period 2000–2024.
Table 2. Summary of environmental variables and flood history metrics calculated for each site during all surveys from field data and the Fisher Water Index applied to LandSat satellite images during the study period 2000–2024.
NameAbbreviationDescription
Long-term flood frequency (2000–2024)Flood frequencyFrequency of flood events across the study period, i.e., number of years where a flood event occurred at each site
Long-term annual flood durationLT_DurationAnnual inundation duration averaged across study period
Years dryDryNumber of consecutive years prior to each survey that a site was dry (i.e., no inundation was recorded)
Consecutive inundationConsecutive floodNumber of consecutive years prior to each survey that a site was inundated
Previous year floodingFP_N, FP_YBinary factor of flooding in the year prior to each survey
Duration inundatedDurationNumber of consecutive days in a water year that inundation occurred at each site
Shrub coverShrubAverage shrub cover in each site calculated from the cover in 10 1 × 1 m quadrats
Burn historyBurnedBinary factor representing site status as burned or not burned in November 2023 determined from mapping of fire extent
SeasonSeasonCategorical variable representing the season during each survey (Table 1)
Table 3. Summary of the taxa observed in understorey vegetation surveys across three floods and six surveys at Narran Lakes. Values in brackets are the native (N) and exotic (E) taxa breakdown of each category.
Table 3. Summary of the taxa observed in understorey vegetation surveys across three floods and six surveys at Narran Lakes. Values in brackets are the native (N) and exotic (E) taxa breakdown of each category.
Survey PeriodNative TaxaExotic TaxaUnknownUnique ObservationsNew ObservationsTotal No. of Taxa
Flood 1 (04–05)5311317
(N: 13, E: 4)		6767
Nov–045011314
(N: 10, E: 4)		6464
May–0522212
(N: 2)		3
(N: 3)		25
Flood 2 (08–09)528416
(N: 13, E: 3)		2064
Nov–08478411
(N: 9, E: 2)		18
(N: 15, E: 3)		59
Apr–0929431
(N: 1)		2
(N: 2)		36
Flood 3 (23–24)5110222
(N: 18, E: 4)		2363
Aug–23399112
(N: 9, E: 3)		15
(N: 11, E: 4)		49
Feb–2428517
(N: 7)		7
(N: 7)		34
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Grieger, R.; Johnston-Bates, J.; Sutton, A.; Capon, S.J. Patterns in Understorey Vegetation of a Semi-Arid Terminal Wetland over 20 Years in Response to Flood and Drought. Diversity 2026, 18, 274. https://doi.org/10.3390/d18050274

AMA Style

Grieger R, Johnston-Bates J, Sutton A, Capon SJ. Patterns in Understorey Vegetation of a Semi-Arid Terminal Wetland over 20 Years in Response to Flood and Drought. Diversity. 2026; 18(5):274. https://doi.org/10.3390/d18050274

Chicago/Turabian Style

Grieger, Rebekah, Jaiden Johnston-Bates, Andres Sutton, and Samantha J. Capon. 2026. "Patterns in Understorey Vegetation of a Semi-Arid Terminal Wetland over 20 Years in Response to Flood and Drought" Diversity 18, no. 5: 274. https://doi.org/10.3390/d18050274

APA Style

Grieger, R., Johnston-Bates, J., Sutton, A., & Capon, S. J. (2026). Patterns in Understorey Vegetation of a Semi-Arid Terminal Wetland over 20 Years in Response to Flood and Drought. Diversity, 18(5), 274. https://doi.org/10.3390/d18050274

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop