Next Article in Journal
Effect of Abiotic Factors on Nectar Quality and Secretion of Two Early Spring Species, Galanthus nivalis L. and Helleborus niger L.
Previous Article in Journal
Indirect Effects of Cattle Trampling on the Structure of Fruit-Feeding Butterfly Assemblages Inhabiting Restinga Forests in Southern Brazil
Previous Article in Special Issue
Linking Perceived Biodiversity and Restorative Benefits in Urban Parks through Place Attachment—A Case Study in Fuzhou, China
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Diversity
Volume 16
Issue 8
10.3390/d16080468
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 thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Finding Isolated Aquatic Habitat: Can Beggars Be Choosers?

by
Danielle M. Husband
† and
Nancy E. McIntyre
*
Department of Biological Sciences, Texas Tech University, Lubbock, TX 79409-3131, USA
*
Author to whom correspondence should be addressed.
†
Current address: U.S. Fish & Wildlife Service, 215 Melody Lane, Wenatchee, WA 98802, USA.
Diversity 2024, 16(8), 468; https://doi.org/10.3390/d16080468 (registering DOI)
Submission received: 4 June 2024 / Revised: 24 July 2024 / Accepted: 1 August 2024 / Published: 3 August 2024
(This article belongs to the Special Issue Diversity in 2024)
Browse Figures
Versions Notes

Abstract

:
In a two-year field study across 58 isolated wetlands in Texas (USA), we examined whether odonate (Insecta: Odonata) assemblages were structured by local environmental filters or instead simply reflected the use of any available water in this semi-arid region. Cluster analysis resolved three wetland groupings based on environmental characteristics (hydroperiod, water chemistry, vegetation); 37 odonate species were detected at these wetlands. The most speciose assemblages occurred at wetlands with longer hydroperiods; these sites also had the most species found at no other wetland type. Ordination plots indicated some filtering with respect to the hydroperiod, but there was only mixed or weak support with respect to other local factors. Because water persistence was the strongest driver maintaining odonate diversity in this region, regardless of water quality or vegetation, beggars cannot be choosers in this system and conservation efforts can focus on water maintenance or supplementation.

1. Introduction

Isolated wetlands provide crucial resources for aquatic insects in otherwise arid and semi-arid environments. Although isolated wetlands are spatially discrete, their biotic communities may be linked via dispersal. Indeed, dispersal is essential for the recolonisation of ephemeral, intermittent, or seasonal wetlands. However, traits such as hydroperiod, water quality, or associated vegetation may act as environmental filters to colonisation because of species-specific tolerances. As a consequence, assemblages of aquatic insects of isolated wetlands are likely subject to regional (dispersal limitation, driven by wetland availability) as well as local (site characteristics fulfilling species-specific niche requirements) influences, but the relative contributions of these factors are still understudied and thus poorly understood [1,2]. Anthropogenic activities are altering hydroperiods, water chemistry, and wetland vegetation, thereby contributing to the decline of aquatic organisms [3,4,5]. The determination of the relative importance of local vs. regional factors affecting diversity is necessary for understanding how aquatic insect assemblages are formed, structured, maintained, and conserved [6,7,8,9]. Therefore, it is imperative to determine which mechanisms are contributing to species diversity to develop appropriate conservation recommendations.
Odonates (dragonflies and damselflies) provide a model taxon for this endeavour. As amphibious insects with aquatic juveniles and flighted adults, assemblages of odonates likely reflect regional and local filters, especially in landscapes characterised by isolated wetlands, such as the Great Plains of North America. The southern portion of the Great Plains in particular is characterised by wetlands that differ in hydrology, such as hydroperiod, and water characteristics, such as salinity; odonates are among their most conspicuous and diverse occupants [10,11]. If Great Plains odonate assemblages are primarily affected by local filters, then differences along axes of hydrology, water chemistry, and/or vegetation should be evident. But given that most of the wetlands of this region are only intermittently inundated, it is possible that “beggars can’t be choosers” (that is, species in a semi-arid climate have few/no choices in aquatic habitat and thus are in no position to pass up any water present) and the assemblages are instead more strongly associated with regional processes (i.e., dispersal limitation from the regional species pool).
There are over 80,000 lentic wetlands in the Southern Great Plains (Figure 1). The most numerous of these are shallow, depressional wetlands called playas; these ephemeral wetlands are dry more often than wet [12]. Fed by runoff, their hydroperiods and water quality are strongly affected by land cover in their watersheds, with many surrounded by row-crop cultivation or by grazed pasture [13]. There are a few (<40) larger wetlands in the Southern Great Plains with freshwater springs and outflow basins that accumulate mineral deposits; these wetlands are regionally known as salinas [14]. Groundwater extraction has lowered the water table to such an extent that many springs are no longer active, leaving dry basins to be solely inundated by rainfall (akin to playas but with saline basins) [14]. In urban areas, stormwater retention ponds provide water year-round from anthropogenic runoff; these are regionally called urban playa lakes. The Southern Great Plains wetlands thus differ—at least in name—on the basis of the hydroperiod (from a gradient of ephemeral to relatively permanent), source of water (precipitation vs. groundwater), water chemistry, as well as surrounding vegetation [15]. However, it is currently unknown whether these wetlands can actually be clustered into discrete, distinguishable types (playa, urban playa lake, salina, or salina with nonfunctioning spring) or whether instead they exist along a more or less continuous gradient of variability. Furthermore, it is unknown whether odonate assemblages are structured by the same environmental variables as the wetlands themselves.
If assemblages of adult odonates group correspondingly to types of wetlands, then (because adults are capable of flight) they must be actively choosing habitat and thus are subject to local-scale (species-specific) filters. If so, then we should be able to make predictions about which taxa should be “filtered out” from wetlands with certain properties. For example, wetlands with short hydroperiods should be unable to support populations of larger-bodied or slower-developing odonate species (i.e., dragonflies), because such wetlands may not contain water for the duration needed for nymphal development for larger or multivoltine species [16,17]. Therefore, we would expect a higher diversity of dragonflies at salinas with functioning springs and urban playa lakes with anthropogenic supplementation of water than at other wetlands. For urban playa lakes, however, which are longer-hydroperiod wetlands, any support for larger, multivoltine species may be countered by various water chemistry variables that are known to negatively affect odonate development, fitness, or survival [18,19,20,21,22]. Thus, we predict that urban water bodies and rural ones surrounded by agriculture (i.e., wetlands affected by anthropogenic runoff) should exhibit lower richness [23] and overall diversity [24] than non-urban, longer-hydroperiod wetlands. Finally, because vegetation is used for oviposition, eclosion, and perching [25,26], odonate diversity is often positively associated with an abundance of littoral vegetation and aquatic macrophytes [26,27,28,29,30,31]. Thus, the removal of vegetation should diminish odonate diversity and conversely, the presence of it should be associated with higher diversity.
These localised environmental factors—hydroperiod, water chemistry, vegetation—may, collectively, group wetlands and thence their odonate assemblages. If no such local filtering is evident, then assemblages are shaped mainly by dispersal, meaning that there would be no strong species-site affiliations because “beggars can’t be choosers”; this should be especially true for dragonflies, which are typically more vagile and can disperse greater distances than can damselflies [32]. In short, odonates colonise isolated wetlands that are variable in this semi-arid landscape. Are the wetlands distinguishable enough in environmental characteristics to support discrete odonate assemblages, or do the assemblages simply reflect the occurrence of water? The answer to this question may influence the kinds of management activities that could be undertaken for conservation of this charismatic and ecologically important group of insects: whereas local environmental traits can be managed for conservation, dispersal limitation is much more difficult (if not impossible) to control.
The majority of the wetlands on the Great Plains are on private property [13] (p. 192), which has constrained our understanding of odonate diversity. Initial surveys focused on roadside ditches and city parks [33], and further studies examined playas [10,11,34,35]. However, there has been no study conducted to date that has assessed odonate diversity beyond these wetlands. The lack of such baseline knowledge hinders the monitoring of the effects of environmental change and establishment of realistic conservation goals [36]. Herein, we address these deficiencies in knowledge to describe how odonate assemblages are structured from local to regional scales. We answered the following research questions: (1) Do the wetlands of the Southern Great Plains form groupings based on hydroperiod, water chemistry, and/or vegetation, or do they exist along continuous axes of variation? (2) Do the odonate assemblages likewise cluster into discrete assemblages on the basis of biotic characteristics such as body size, development time, and sensitivity to various aspects of water chemistry? (3) Do the assemblage groupings correspond to wetland groupings? If wetlands do exhibit characteristics that cause them to be separated into groups, then habitat selection by odonates is possible (making them “choosers”). If the odonate assemblages do not reflect wetland groupings, however, then odonates are taking advantage of whatever water is present in this semi-arid area (making them “beggars”).

2. Materials and Methods

We tallied the numbers of odonate species present at each of the 58 wetlands in Texas across two summers. Local environmental variables were measured at each wetland. We then determined whether wetland sites formed discrete clusters or instead were continuously distributed along gradients of these variables. Finally, odonate occurrence was examined to determine whether there was assemblage structuring with respect to environmental variables. Our methodological approach is one that can be adopted for any taxon. We used presence/absence (occurrence) data, but abundance data could also be used. Abundance data are notoriously difficult to obtain for adult odonates, usually necessitating collection of specimens [32], but occurrence can serve as a proxy for abundance [37].

2.1. Study Sites

We surveyed 58 wetlands in Texas in summers of 2020–2021 (21 playas [17 surrounded by row-crop agriculture, 4 by grazed pastures], 26 urban playa lakes, 4 salinas with functioning springs, and 7 salinas with dry springs; Figure 2). Surveyed wetlands were separated by at least 500 m. Survey locations were situated on the perimeter of each wetland basin (i.e., the littoral zone), as this is where odonate activity is concentrated [32]. Environmental variables and odonates were measured only at sites that held water; this differed between years. Thus, in summer 2020, we sampled 27 inundated wetlands; in summer 2021 (wetter than 2020), we sampled 48 inundated wetlands (Table 1). Twenty-six wetlands were able to be sampled in both years (Table 1).

2.2. Local Environmental Variables

At each wetland, we measured local environmental variables (hydroperiod, presence and percent coverage of emergent vegetation, presence of floating macrophytes in the littoral zone, and an array of water chemistry variables; Table 2). For hydroperiod, we classified wetlands as being inundated for either a long (essentially permanent; N = 30 wetlands) or short (dry more often than wet; N = 28) period of time based on examining a time series of imagery from Google Earth Pro (1985–2021): categorization of long hydroperiod was based on visual evidence of water on >50% of imagery dates. Google Earth Pro imagery resolution is typically <0.5 m but varies depending on the satellite used, date, and other factors. We examined all dates provided by Google Earth Pro, even those occurring during the winter, as some dragonflies have a nymphal stage that lasts for as many as five years [32], meaning that the presence or absence of water on even winter dates is important in being able to support odonate diversity. We examined an average of 16.86 (range: 9–32) dates per wetland. Wetlands classified as long hydroperiod were wet an average of 96.6% of dates examined (range: 83–100%), whereas short-hydroperiod wetlands were wet only on 16.4% of dates (range: 0–30%); there were no wetlands close to the 50% cutoff. Thus, our binary hydroperiod classification accurately captured wet/dry patterns in this area.
We visually estimated percentage of shoreline (in increments of 1%) with vegetation during each sampling visit. Ten water chemistry variables that have been used in other odonate studies were measured [19,23,24,38,39,40,41]; conductivity (μS), salinity (ppm), pH, and total dissolved solids (TDS ppm) were measured in situ with a Waterproof ExStik® II (Extech Instruments, Boston, MA, USA) whereas the other water chemistry variables were measured in the lab. Two 50 mL water samples were collected at those wetlands with water present for some wetlands, only one sample could be collected if they were inundated in one summer but not the other). For wetlands with active springs (i.e., salinas), water samples were collected from the spring(s) because water flow dried before reaching the basin proper and odonate activity was concentrated at the springs. For wetlands that had multiple springs, two water samples were collected from each spring and water chemistry values were averaged. Water samples were processed at the Department of Biological Sciences at Texas Tech University on the same day they were collected. We measured nitrate (mg/L NO3), nitrite (mg/L NO2-N), sulphate (mg/L SO4), ammonia (mg/L NH3-N), and phosphate (mg/L PO43−) with a Drel 2400 (Hach, Loveland, Colorado, USA) complete water quality kit, whereas turbidity (NTU) was evaluated with a Hach 2100P Turbidimeter (Hach, Loveland, CO, USA).

2.3. Odonate Data

Odonate occurrence data were recorded between 1 June and 30 September 2020 and 2021 on days above 23 °C with clear to moderately clear skies and low wind (<35 km/hr) between 0900 and 1400, following a protocol similar to [42]. Data were recorded using an Esri Survey123 mobile application designed by D.M. Husband. All adult odonates were visually identified at three survey points spaced at least 50 m apart per wetland per visit. Estimating adult odonate richness at a given site requires a balance among sampling duration, frequency, and site replication [43]. There are multiple ways of assessing odonate richness at a given site, some with standardised protocols (e.g., point counts or belt transects that may be standardised in terms of time spent surveying or be area-weighted), whereas others are opportunistic [see, e.g., 7, 10, 35, 42 for some examples of these methods]. Surveys may be stationary (e.g., point counts) or moving (e.g., transects). Survey duration must be weighed against lost opportunity costs in sampling other locations, so the point of diminishing returns must be identified, but this is rarely conducted empirically [43].
Because our focal region was >41,000 km2 in area and contained hundreds of wetlands (Figure 2), we decided to try to maximise the number of sites surveyed so as to be able to capture variability among them. We therefore used a combination of stationary point counts but had three such counts at different locations at each wetland; this allowed us to potentially encounter species that were localised due to territory or microhabitat. Our estimated adult odonate richness was thus a function of the sampling duration, number of site visits per summer, and travel time among wetlands that were hundreds of kilometres apart. We used a fixed-time point-count protocol, with data pooled across three survey point locations per wetland. Surveys lasted 15 min (5 min per survey point) and each wetland was surveyed once during early summer and once during late summer. Twenty-one sites were surveyed only in either 2020 or 2021 due to land access permission or lack of water, meaning that the number of sampling visits per site over the duration of our study ranged from 2 to 4. During these visits, odonate species identifications were made from visualisations, photos, and collected specimens (voucher specimens are housed in the Department of Biological Sciences at Texas Tech University).
A longer sampling duration would be needed for lentic systems and species that patrol long stretches of streams, which was not the case in our focal area. As long as the sampling duration is constant across sites, then sites can be compared equitably. Presumably, however, there is some threshold in the amount of time that one must spend surveying a site to obtain a reasonable approximation of species richness: 10 s counts would be too short to provide such data, even if conducted equitably across sites. To determine whether pooling three 5 min samples per wetland per visit adequately represented odonate richness, four of our top seven most speciose wetlands (U1, U5, U0, U21) were visited by N.E. McIntyre in July 2024. Odonate occurrence data were taken at 5 min intervals for 60 min at two locations per wetland, separated by at least 50 m. We determined how many species out of the total seen in 60 min/location and 120 min/wetland were seen in the first five minutes, and also whether any species missed in the first five minutes were seen in the first five minutes at the second location per wetland. Finally, species accumulation curves were used to determine what survey duration was needed to asymptotically represent odonate richness at each survey location and at each wetland as a whole.
Adults were our focus for two crucial reasons. First and foremost, this is the dispersive stage (nymphs are strictly aquatic) and so local adult diversity reflects the importance of dispersal potential (a regional filter). Although some dragonfly species are known to be migratory and capable of flying >100 km, most odonates do not disperse very far from their natal sites (typically <500 m [32,44,45,46]), and we sampled wetlands that were spaced at least 500 m apart. Secondly, adults are far more readily and accurately identified to species compared to other life stages and thus are commonly used in environmental monitoring [46,47,48,49]. Although nymphs or exuviae provide stronger evidence of population establishment at a wetland than do adults [50,51], adult odonate presence can be used to indicate breeding sites [52,53,54]. Moreover, there simply are no taxonomic keys to exuviae at all, nor for nymphs of damselflies for this region, and even keys for dragonfly nymphs require F-0 instars [55]. Finally, there is evidence that odonate nymphs and adults exhibit similar patterns of occurrence, richness, and abundance [56]. Although presence of an adult at a given wetland does not guarantee that it had developed in and emerged from that wetland, its presence still indicates use of that wetland (if only for navigation, hunting, etc.). Therefore, our assemblages reflect overall local diversity. Although this could possibly lead to overestimation of the importance of regional (dispersal) factors in driving odonate diversity in wetlands of the Southern Great Plains, it is the adults who have the potential to differentiate wetlands and begin the process of differential community assembly.
To test our predictions related to local environmental filters, we used the primary literature and field guide descriptions to categorise adult odonates a priori according to their development time (long or short), chemical sensitivity, and salt tolerance (Table S1). This allowed us to predict which taxa should be present/absent from a given wetland or wetland cluster. If wetlands clustered on the basis of one or more environmental traits, then these descriptions would potentially be useful in explaining species presences/absences from a given wetland or wetland cluster.

2.4. Analyses

All analyses were conducted in RStudio 2023.06.1 [57]. Species accumulation curves were plotted using the collector’s method in package vegan [58]. A k-means cluster analysis of wetlands that contained water was performed with package cluster [59]. This iterative process seeks to determine the optimal number of non-overlapping clusters present in the data [60]; we evaluated the occurrence of 2–10 possible clusters of wetlands, based on hydroperiod, water chemistry, and vegetation. Prior to analysis, environmental data were transformed (non-percent values by Log10 and percent values by arcsine square root), and variables that were highly correlated (Spearman r > |0.8|) were removed from all analyses. (This removed conductivity, NH3, and total dissolved solids from further analyses.) Finally, environmental data were averaged by wetland and relativized (scaled and centred) prior to analyses. A Principal Component Analysis (PCA) was also performed on the wetland data to determine how much variation in the data could be explained, which provided a form of corroboration of our cluster analysis. Jenks Natural Breaks was used to identify relatively low, medium, and high values of average water chemistry values and of vegetation cover by wetland and cluster. Resolved wetland groupings were then examined with respect to their odonate assemblages via redundancy analysis (RDA) in package vegan, using the Hellinger transformation on our odonate occurrence data, which standardises relative abundances across species for ordination [61]. The odonate assemblage data were also examined via cluster analysis to determine whether they followed similar patterns as the wetlands. For the numerical variables (i.e., all variables other than the categorical variable of hydroperiod), a permutational multivariate analysis of variance (PERMANOVA) was used to identify the environmental variables responsible for differences in assemblage structure. Finally, a “best subset of environmental variables” (BIOENV) analysis with maximum rank correlation with community dissimilarities was applied to normalised abiotic variables each year to determine which variables produced the highest correlation with assemblage richness, providing additional support as to which measured variables structure the odonate assemblage of the Southern Great Plains. All data and analytical code are available at https://github.com/D-Husband/Odonata_Panhandle-/tree/main (accessed on 3 June 2024).

3. Results

In our system, five-minute point count surveys accounted for 84% of all species seen within a full hour at each sampling location (range: 60–100%) and 84% of all species seen at each wetland over a 120 min span (range: 67–93%), with 50–67% of species missed within the first 5 min at one sampling location observed within 5 min at the second sampling location. This is reflected in the relatively flat, asymptotic species accumulation curves at each wetland (Figure 3). These findings indicate that our pooled five-minute surveys did an acceptable job at representing odonate richness at each wetland (indeed, all species observed within an hour were observed within the first five minutes at three of the sampling locations). Moreover, no new species were encountered at any locality after 40 min (Figure 3).
Odonate assemblages likewise varied across wetlands of the Southern High Plains. Twenty-seven species were recorded in 2020; 35 were found in summer 2021. These differences are likely due to the fact that fewer wetlands held water in 2020 than in 2021 (Table 1). Over both summers, 37 odonate species were detected, accounting for 40.6% of the total odonate species diversity known from this region based on data from OdonataCentral.org, which includes species from flowing-water and large lake habitats that we did not examine; when lotic and large lake species are excluded, then our samples captured ~50% of all odonate species ever encountered in our area since records began being kept in the 1990s. Although the odonate species grouped into three clusters, they did not coincide with the three wetland clusters (Table 3, Figure 5 and Figure S6): instead, there were 17–32 odonate species per wetland cluster, with 12 species occurring at only a single wetland; nine species occurred at multiple wetlands but within a single cluster (Table 3 and Table 4, Figure 5). The highest richness was associated with clusters 1 and 2, which comprised long-hydroperiod, alkaline wetlands (Table 2). As expected, proportionally more damselflies were associated with a single site (5/14 species = 35.7%) than dragonflies (7/23 species = 30%) (Table 3).
Most species were relatively rare, occurring at only one or a few wetlands, but one species (Enallagma civile) occurred at every wetland. The highest species richness was found at wetlands with the longest hydroperiods, particularly salinas with functioning springs, followed by urban playa lakes; the most unique species (i.e., those species found at only one wetland) were also at salinas. Wetlands with short hydroperiods did not include any unique species (i.e., species found at no other site).
Redundancy analysis on the Hellinger-transformed odonate data revealed that the environmental variables could explain only 19.33% of the variance in the odonate data (80.66% was unconstrained by the environmental variables). PERMANOVA indicated that only percent vegetation (p = 0.021) and sulphate concentration (p = 0.013) drove patterns in the odonate assemblages among wetlands; the remaining numerical variables (nitrate, nitrite, pH, phosphate, salinity, and turbidity) did not significantly contribute to the RDA model (p > 0.05). However, whereas there was clear separation in the odonate assemblages by hydroperiod, that was not the case for vegetation nor sulphate, which did not separate out into clear groupings (Figure 6). High sulphate levels were associated with several wetlands belonging to clusters 1 and 2; high vegetation levels likewise occurred at several wetlands belonging to all three resolved wetland clusters (Table 2).
BIOENV analysis determined that odonate richness at sampled wetlands was significantly correlated with percent vegetation, albeit weakly (r = 0.1350); all other numerical variables were not significant. As predicted, we found the highest number of dragonfly species at salinas with functioning springs and at urban playa lakes with anthropogenic supplementation of water (i.e., longest-hydroperiod wetlands). We did not find any consistent effects with respect to anthropogenic runoff/compromised water quality, as there was higher diversity at urban playa lakes but not at playas surrounded by agriculture. Thus, hydroperiod may be overriding localised selection separation.
The a priori species descriptions (Table S1) helped explain assemblage-level patterns among the three resolved wetland clusters. For example, the largest-bodied univoltine species (e.g., Anax junius, Phanogomphus militaris, Rhionaeschna multicolor) were found primarily or exclusively at the wetlands with the longest hydroperiods (Table S1 and Table 3). Most of the known salt-tolerant species (e.g., Brachymesia gravida, Erythemis simplicicollis, Lestes alacer, Libellula composita) were found primarily or exclusively at salinas or at urban playa lakes with relatively high salinity values (Table S1 and Table 3). Finally, all of the species with known tolerance of anthropogenic chemicals were damselflies. Most of these species (e.g., Argia alberta, Argia immunda, Ischnura hastata, Ischnura ramburii) were exclusively found at urban sites or in cluster 2, which was characterised by moderate-to-high concentrations of most of the anions we examined (Table S1 and Table 3).

4. Discussion

Although we were able to resolve three clusters of wetlands based on their environmental characteristics and three clusters of odonate species co-occurrences, the wetland and odonate clusters were not concordant. The lack of overlap between wetland and odonate clusters, combined with a positive association seen with hydroperiod but only mixed or weak support with respect to water chemistry variables and vegetation, indicates that regional rather than local filters are more likely to be of relatively greater importance in structuring the odonate assemblages of the wetlands of the Southern Great Plains. Of course, both regional and local factors are present, and presumably some species are more strongly influenced by one or the other. As we had predicted, we found a higher diversity of dragonflies and damselflies at salinas with functioning springs and at urban playa lakes with anthropogenic supplementation of water than at other wetlands. With respect to water chemistry, wetlands surrounded by urban development and agriculture did not have lower damselfly richness or overall diversity but did have lower dragonfly richness. Finally, in terms of surrounding vegetation, agricultural playas did have lower diversity, but urbanised ones did not. We found that the wetlands of the Southern Great Plains formed groupings, primarily based on hydroperiod, and that the odonate assemblages likewise clustered into discrete assemblages. However, the assemblage groupings did not correspond to wetland groupings, meaning that the odonates are taking advantage of whatever water is present (making them “beggars” rather than “choosers”). Thus, the environmental variables we measured did not explain the majority of the variation seen in the odonate assemblage data (R2 = 0.1933). The lack of a strong effect could also have been due to the fact that we performed a very conservative statistical analysis, with a transformation that downweighs the influence of rare species (which made up the majority of diversity in any assemblage, including ours) [62].
Eight species (Anax junius, Libellula pulchella, Pantala flavescens, P. hymenaea, Sympetrum corruptum, Tramea lacerata, T. onusta, Enallagma civile) represented a large proportion of the diversity observed at wetlands with relatively short hydroperiods. We found more species and more unique species at longer-hydroperiod wetlands. This result supports the idea that odonate assemblages at wetlands with relatively longer hydroperiods differ from those with shorter hydroperiods, with longer hydroperiods being essential for many odonate species in this semi-arid region.
Hydroperiod structures assemblages through a variety of mechanisms [63]. For example, wetlands with longer hydroperiods can support more complex food webs [64]. These wetlands may also support more predators like fish, however, which negatively affect odonates [42]. The wetlands of our focal area are naturally fishless [13], although some urban parks in our focal region are stocked with fish [64]. Anecdotally, we observed fewer odonates at some wetlands with stocked fish but not others.
Urbanisation is associated with lower odonate richness in some areas [65,66], but we found high diversity at urban playa lakes (second only to salinas). This pattern (higher odonate richness at urban wetlands than at rural ones) has been observed in our region [67] and elsewhere [37,68]. In our study, this pattern could be because urbanisation in this grassland biome is associated with the creation of greater habitat heterogeneity than would otherwise be found, primarily through afforestation; habitat heterogeneity is positively associated with greater odonate diversity [2,68]. Urban playa lakes, being used in stormwater drainage, had relatively long hydroperiods, but the presence of water is not the only factor that could be responsible for the high species richness that we saw at those wetlands. Distinguishing how habitat heterogeneity and hydroperiod affect odonate diversity would require additional data, but because these environmental variables are associated with each other and are at work simultaneously, it may be impossible to tease apart which factor is the single most influential one.
The odonate assemblages at different wetlands were similar (Table 3), but some differences in environmental conditions were noted by year that could have potentially affected the regional species pool. Because only ~62% of the annual average precipitation occurred in 2020, that summer was the 11th driest on record [68]; many playas were completely dry that year. Summer 2021 received more rainfall than in 2020, allowing more wetlands to be surveyed that year. Because they could only be surveyed in 2021 (since they were dry in 2020), playas had the fewest observed species (N = 9) and contained many of the species that were found at all or nearly all other wetlands. The ephemeral nature of playas means it is not surprising that their assemblage largely consisted of regionally common species and genera known to be migratory (e.g., Tramea, Sympetrum). Twelve species were unique to salinas; these included species known to be salt-tolerant (e.g., Libellula composita) [69]. Other species like Phanogomphus militaris, Dythemis fugax, and several Ischnura species that were detected only at salinas may be salt-tolerant, but further investigation is needed on these species. Salinity is known to affect the abundance of many aquatic invertebrates [70,71,72]. In our study, salinity was found to influence community structure, but odonate assemblages at urban playa lakes and former salinas (the wetlands where salinity values were highest) were dissimilar, indicating that factors other than salinity drive odonate diversity in this largely freshwater area. Odonate richness was low at most former salinas where the highest salinity was measured. Similar patterns have been observed in wetlands of the Prairie Pothole Region where brine inputs from oil and gas industries have increased salinity in otherwise freshwater wetlands [73]. However, odonate richness at urban playa lakes did not reflect the same pattern of low diversity at high salinity sites, which suggests that another factor may be driving urban odonate diversity.
Although the RDA-recovered sulphate concentration ais a relatively important variable, no species’ occurrence was associated with it nor any of the other water chemistry variables we measured, possibly because there was extreme variation across wetlands. For example, sulphate and salinity were two variables that significantly differed among sampled wetlands in both years. Sulphate naturally occurs in groundwater, so its higher concentration at salinas and former salinas is not surprising given that these are or were spring-fed wetlands [14]. Sulphate presence in urban wetlands may be due to the fact that this region’s alkaline soils cause iron deficiencies in many plants, which is treated with applications of sulphate [74,75,76]. However, there have been no studies that have examined the effects of high sulphate levels on odonates. This variability in water chemistry across wetlands means it would be difficult for the local sorting of colonisers to occur. Long-term monitoring of wetlands and odonates in this region would be needed to fully understand the effects of environmental variation on habitat selection and biotic assembly. Long-term studies are sorely needed in general to document dynamics in odonate populations and to understand the patterns of their abundance and distribution [77].
Each method for surveying adult odonates has its advantages and disadvantages. For example, our method of pooling data from relatively brief stationary point counts taken at multiple locations per wetland was expeditious and effective at representing the total species richness in our focal system. This may have been because we had a relatively depauperate regional species pool (low richness and low abundances) at mostly small and open wetlands. A greater time investment would likely be needed in larger or more structurally complex wetlands with more diverse assemblages (particularly if there is a high degree of endemism possible), although this would need to be tempered if trying to estimate abundances, as the chance of double counting an individual increases with count duration. With respect to richness as in this study, however, long surveys may be needed to encounter more wary species or species that patrol very large areas and thus only happen by a given spot infrequently.

5. Conclusions

Although categorisations such as “playa”, “salina”, etc. are used extensively in the literature (with respect to odonates as well as on other topics, e.g., [10,11,12,13,14,15,34,35,37,67,78]), our findings indicate that these labels are not always biologically relevant. With respect to odonates, not only did the wetlands not cluster into groupings that corresponded to their labels, but the clusters did also not parse out along various environmental axes, save hydroperiod. Thus, the simple presence of water, regardless of its chemistry, vegetation, etc., was the primary factor dictating odonate occurrence most strongly in our focal region. Because water persistence was the strongest driver maintaining odonate diversity, conservation efforts can focus on water maintenance or supplementation. Most of the region’s conservation programmes work one-on-one with landowners, given that most of the region’s wetlands occur on private property [13], and there is otherwise a lack of legislative protection for the wetlands of the Southern Great Plains. Most of the conservation efforts are incentive-based. For example, the U.S. Department of Agriculture’s Conservation Reserve Program and non-profit organisations such as the Playa Lakes Joint Venture provide guidance and resources for landowners [13,78,79]. Of especial concern are salinas, given how many odonate species were found nowhere else (Table 3), with estimates of <10 with functioning springs remaining in Texas [14,80].
Because most of the wetlands of this region dry up frequently and for long durations (some are dry for years at a time), they may act as “ecological traps” with hidden costs to persistence and/or fitness [81]. Such may be the case for at least some urban playa lakes, although some recent studies have found that urban wetlands can support a diverse odonate community despite changes in macrophyte cover and various indices of water quality [2,66,82,83,84]. In our study area, the urban parks are mowed to the urban playa lakes’ edges, which eliminates vegetation for emergence and perching. Neither the state nor local municipalities appear to test water quality at urban playa lakes, and there are no programmes on how water quality issues at urban playas are addressed and remediated. Because urban playa lakes are consistently wet from anthropogenic sources of water, odonates could potentially serve as a model taxon to monitor these sites. Urban parks also present opportunities for public engagement and education on the importance of these charismatic organisms and of wetlands.

Supplementary Materials

The following supporting information can be downloaded at https://www.mdpi.com/article/10.3390/d16080468/s1: Table S1: Categorisation of the odonate species we encountered at 58 wetlands in Texas in 2020–2021 in terms of their ontogenetic development time (L = long, S = short), chemical sensitivity (Y = yes, N = no), and salt tolerance (Y = yes, N = no). Cells with “--“ indicate no information is known about that species’ salinity tolerance. Species are listed in alphabetical order starting with dragonflies, followed by damselflies [30,32,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102]. Figure S1: PCA biplot illustrating how our wetlands are structured by environmental variables. Wetlands are indicated by site codes from Table 1; their positions indicate the scores of the observations on the first two principal components. Environmental variables are indicated by red vectors, representing the coefficients of the variables on the first two principal components; vector length is positively associated with strength of the association. Proximity of a wetland to an environmental vector indicates a positive association. Figure S2: Scree plot illustrating the percent variance explained by each principal component (Dimension). Figure S3: Environmental variables most strongly influencing principal component 1. Figure S4: Environmental variables most strongly influencing principal component 2. Figure S5: Environmental variables most strongly influencing principal component 3. Figure S6: Three non-overlapping clusters were found for wetlands (left) and odonate assemblages (right).

Author Contributions

Conceptualization, D.M.H. and N.E.M.; methodology, D.M.H.; software, D.M.H.; validation, D.M.H.; formal analysis, D.M.H.; investigation, D.M.H.; resources, N.E.M.; data curation, D.M.H. and N.E.M.; writing—original draft preparation, D.M.H.; writing—review and editing, N.E.M.; visualisation, D.M.H.; supervision, N.E.M.; project administration, N.E.M.; funding acquisition, N.E.M. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Data Availability Statement

Data and analytical code are available at https://github.com/D-Husband/Odonata_Panhandle-/tree/main (accessed on 3 June 2024).

Acknowledgments

We express our gratitude to the landowners who allowed us to survey on their properties. We thank Austin Biddy and Courtney Miller for assistance in the field and Natasja van Gestel for statistical advice. Justin Dawsey, Hannah Girgente, Joe Girgente, Kerry Griffis-Kyle, Tigga Kingston, and five anonymous reviewers provided comments on earlier manuscript drafts.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Galbraith, H.S.; Vaughn, C.C.; Meier, C.K. Environmental variables interact across spatial scales to structure Trichopteran assemblages in Ouachita mountain rivers. Hydrobiologia 2008, 596, 401–411. [Google Scholar] [CrossRef]
  2. Thornhill, I.; Batty, L.; Death, R.G.; Friberg, N.R.; Ledger, M.E. Local and landscape scale determinants of macroinvertebrate assemblages and their conservation value in ponds across an urban land-use gradient. Biodiv. Conserv. 2017, 26, 1065–1086. [Google Scholar] [CrossRef] [PubMed]
  3. Collen, B.; Whitton, F.; Dyer, E.E.; Baillie, J.E.M.; Cumberlidge, N.; Darwall, W.R.T.; Pollock, C.; Richman, N.I.; Soulsby, A.-M.; Böhm, M. Global patterns of freshwater species diversity, threat and endemism. Glob. Ecol. Biogeogr. 2014, 23, 40–51. [Google Scholar] [CrossRef]
  4. Collier, K.J.; Probert, P.K.; Jeffries, M. Conservation of aquatic invertebrates: Concerns, challenges, conundrums. Aquat. Conserv. Mar. Freshw. Ecosyst. 2016, 26, 817–837. [Google Scholar] [CrossRef]
  5. Reid, A.J.; Carlson, A.K.; Creed, I.F.; Eliason, E.J.; Gell, P.A.; Johnson, P.T.J.; Kidd, K.A.; MacCormack, T.J.; Olden, J.D.; Ormerod, S.J.; et al. Emerging threats and persistent conservation challenges for freshwater biodiversity. Biol. Rev. 2019, 94, 849–873. [Google Scholar] [CrossRef]
  6. Heino, J.; Mykrä, H. Control of stream insect assemblages: Roles of spatial configuration and local environmental factors. Ecol. Entomol. 2008, 33, 614–622. [Google Scholar] [CrossRef]
  7. Juen, L.; De Marco, P., Jr. Odonate biodiversity in terra-firme forest streamlets in Central Amazonia: On the relative effects of neutral and niche drivers at small geographical extents. Insect Conserv. Div. 2011, 4, 265–274. [Google Scholar] [CrossRef]
  8. Arribas, P.; Velasco, J.; Abellán, P.; Sánchez-Fernández, D.; Andújar, C.; Calosi, P.; Millán, A.; Ribera, I.; Bilton, D.T. Dispersal ability rather than ecological tolerance drives differences in range size between lentic and lotic water beetles (Coleoptera: Hydrophilidae). J. Biogeogr. 2012, 39, 984–994. [Google Scholar] [CrossRef]
  9. Bispo, P.C.; Zanini Branco, C.C.; Carneiro Novaes, M.; Costa, L.S.M.; Yokoyama, E.; Barfield, M.; Holt, R.D. Partitioning and drivers of aquatic insect beta diversity in mountain streams. Freshw. Sci. 2021, 40, 644–658. [Google Scholar] [CrossRef]
  10. Reece, B.A.; McIntyre, N.E. Odonata of playas in the Southern High Plains, Texas. Southwest. Nat. 2009, 54, 96–99. [Google Scholar] [CrossRef]
  11. Reece, B.A.; McIntyre, N.E. Community assemblage patterns of odonates inhabiting a wetland complex influenced by anthropogenic disturbance. Insect Conserv. Div. 2009, 2, 73–80. [Google Scholar] [CrossRef]
  12. McIntyre, N.E.; Collins, S.D.; Heintzman, L.J.; Starr, S.M.; van Gestel, N. The challenge of assaying landscape connectivity in a changing world: A 17-year case study in the Southern Great Plains (USA) playa network. Ecol. Indic. 2018, 91, 607–616. [Google Scholar] [CrossRef]
  13. Smith, L. Playas of the Great Plains; University of Texas Press: Austin, TX, USA, 2003. [Google Scholar]
  14. Heintzman, L.J.; Starr, S.M.; Mulligan, K.R.; Barbato, L.S.; McIntyre, N.E. Using satellite imagery to examine the relationship between surface-water dynamics of the salt lakes of western Texas and Ogallala Aquifer depletion. Wetlands 2017, 37, 1055–1065. [Google Scholar] [CrossRef]
  15. Husband, D.M. Odonate Assemblages of Four Wetland Types Varying in Salinity, Hydroperiod, and Water Chemistry in the Texas Panhandle. Master’s Thesis, Texas Tech University, Lubbock, TX, USA, 2022. [Google Scholar]
  16. Crumrine, P.W.; Switzer, P.V.; Crowley, P.H. Structure and dynamics of odonate communities: Accessing habitat, responding to risk, and enabling reproduction. In Dragonflies and Damselflies: Model Organisms for Ecological and Evolutionary Research, 1st ed.; Córdoba-Aguilar, A., Ed.; Oxford University Press: New York, NY, USA, 2008; pp. 21–38. [Google Scholar]
  17. Pires, M.M.; Stenert, C.; Maltchik, L. Partitioning beta-diversity through different pond hydroperiod lengths reveals predominance of nestedness in assemblages of immature odonates. Entomol. Sci. 2017, 20, 318–326. [Google Scholar] [CrossRef]
  18. Chang, X.; Zhai, B.; Wang, M.; Wang, B. Relationship between exposure to an insecticide and fluctuating asymmetry in a damselfly (Odonata, Coenagriidae). Hydrobiologia 2007, 586, 213–220. [Google Scholar] [CrossRef]
  19. Al-Shami, S.A.; Hisamuddin, S.N.; Rawi, C.S.M.; Abdul, N.H.; Ahmad, A.H. Developmental instability in Odonata larvae in relation to water quality of Serdang River, Kedah, Malaysia. Life Sci. J. 2014, 11, 152–159. [Google Scholar]
  20. Stoks, R.; Debecker, S.; Van, K.D.; Janssens, L. Integrating ecology and evolution in aquatic toxicology: Insights from damselflies. Freshw. Sci. 2015, 34, 1032–1039. [Google Scholar] [CrossRef]
  21. Mangahas, R.S.; Murry, R.L.; McCauley, S.J. Chronic exposure to high concentrations of road salt decreases immune response of dragonfly larvae. Front. Ecol. Evol. 2019, 7, 376. [Google Scholar] [CrossRef]
  22. Beketov, M.A. Ammonia toxicity to larvae of Erythromma najas (Hansemann), Lestes sponsa (Hansemann) and Sympetrum flaveolum (Linnaeus) (Zygoptera: Coenagrionidae, Lestidae; Anisoptera: Libellulidae). Odonatologica 2002, 31, 297–304. [Google Scholar]
  23. Osborn, R. Odonata as indicators of habitat quality at lakes in Louisiana, United States. Odonatologica 2005, 34, 259–270. [Google Scholar]
  24. Ishak, M.; Norhisham, A.R.; Thomas, S.M.; Nurhidayu, S.; Ghazali, A.; Azhar, B. Physicochemical properties as driver of odonata diversity in oil palm waterways. Front. For. Glob. Chang. 2021, 4, 613064. [Google Scholar] [CrossRef]
  25. Rembsurg, A.J.; Turner, M.G. Aquatic and terrestrial drivers of dragonfly (Odonata) assemblages within and among north-temperate lakes. J. N. Am. Benthol. Soc. 2009, 28, 44–56. [Google Scholar]
  26. Brito, J.S.; Michelin, T.S.; Juen, L. Aquatic macrophytes are important substrates for Libellulidae (Odonata) larvae and adults. Limnology 2021, 22, 139–149. [Google Scholar] [CrossRef]
  27. Schindler, M.; Fesl, C.; Chovanec, A. Dragonfly associations (Insecta: Odonata) in relation to habitat variables: A multivariate approach. Hydrobiologia 2003, 497, 169–180. [Google Scholar] [CrossRef]
  28. Janssen, A.; Hunger, H.; Konold, W.; Pufal, G.; Staab, M. Simple pond restoration measures increase dragonfly (Insecta: Odonata) diversity. Biodiv. Conserv. 2018, 27, 2311–2325. [Google Scholar] [CrossRef]
  29. Perron, M.A.C.; Richmond, I.C.; Pick, F.R. Plants, water quality and land cover as drivers of Odonata assemblages in urban ponds. Sci. Total Environ. 2021, 773, 145467. [Google Scholar] [CrossRef]
  30. Worthen, W.B.; Fravel, R.K.; Horne, C.P. Downstream changes in odonate (Insecta: Odonata) communities along a suburban to urban gradient: Untangling natural and anthropogenic effects. Insects 2021, 12, 201. [Google Scholar] [CrossRef] [PubMed]
  31. Jeanmougin, M.; Leprieur, F.; Loïs, G.; Clergeau, P. Fine-scale urbanization affects Odonata species diversity in ponds of a megacity (Paris, France). Acta Oecol. 2014, 59, 26–34. [Google Scholar] [CrossRef]
  32. Corbet, P.S. Dragonflies: Behavior and Ecology of Odonata; Comstock Publishing Associates: New York, NY, USA, 1999. [Google Scholar]
  33. Abbott, J.C. Distribution of dragonflies and damselflies (Odonata) in Texas. Trans. Am. Entomol. Soc. 2001, 127, 199–228. [Google Scholar]
  34. Hernandez, K.M.; Reece, B.A.; McIntyre, N.E. Effects of anthropogenic land use on Odonata in playas of the Southern High Plains. West. N. Am. Nat. 2006, 66, 273–278. [Google Scholar] [CrossRef]
  35. Craig, C.N.; Reece, B.A.; McIntyre, N.E. Nestedness in playa odonates as a function of area and surrounding land-use. Wetlands 2008, 28, 995–1003. [Google Scholar] [CrossRef]
  36. Oertli, B.; Parris, K.M. Review: Toward management of urban ponds for freshwater biodiversity. Ecosphere 2019, 10, e02810. [Google Scholar] [CrossRef]
  37. Callaghan, C.T.; Borda-de-Água, L.; van Klink, R.; Rozzi, R.; Pereira, H.M. Unveiling global species abundance distributions. Nature Ecol. Evol. 2023, 7, 1600–1609. [Google Scholar] [CrossRef] [PubMed]
  38. Holtmann, L.; Juchem, M.; Brüggeshemke, J.; Möhlmeyer, A.; Fartmann, T. Stormwater ponds promote dragonfly (Odonata) species richness and density in urban areas. Ecol. Eng. 2018, 118, 1–11. [Google Scholar] [CrossRef]
  39. Perron, M.A.C.; Pick, F.R. Stormwater ponds as habitat for Odonata in urban areas: The importance of obligate wetland plant species. Biodiv. Conserv. 2020, 29, 913–931. [Google Scholar] [CrossRef]
  40. Vilenica, M.; Brigić, A.; Štih Koren, A.; Koren, T.; Sertić Perić, M.; Schmidt, B.; Bužan, T.; Gottstein, S. Odonata assemblages in urban semi-natural wetlands. Insects 2024, 15, 207. [Google Scholar] [CrossRef] [PubMed]
  41. Maciel Barbosa dos Santos, F.; Juen, L.; Cajaiba, R.L.; Pereira de Sousa, J.R. Distribution of the Odonata assemblages along an environmental gradient in the streams of the legal Amazonia region in western Maranhão (Brazil). J. Insect Conserv. 2024. [Google Scholar] [CrossRef]
  42. Cezário, R.R.; Pena Firme, P.; Pestana, G.C.; Vilela, D.S.; June, L.; Cordero-Rivera, A.; Guillermo, R. Sampling methods for dragonflies and damselflies. In Measuring Arthropod Biodiversity; Santos, J.C., Fernandes, G.W., Eds.; Springer: Cham, Switzerland, 2021; pp. 223–240. [Google Scholar]
  43. Bried, J.T.; Hager, B.J.; Hunt, P.D.; Fox, J.N.; Jensen, H.J.; Vowels, K.M. Bias of reduced-effort community surveys for adult Odonata of lentic waters. Insect Conserv. Div. 2011, 5, 213–222. [Google Scholar] [CrossRef]
  44. McPeek, M.A. Determination of species composition in the Enallagma damselfly assemblages of permanent lakes. Ecology 1990, 71, 83–98. [Google Scholar] [CrossRef]
  45. Rouquette, J.R.; Thompson, D.J. Patterns of movement and dispersal in an endangered damselfly and the consequences for its management. J. Appl. Ecol. 2007, 44, 692–701. [Google Scholar] [CrossRef]
  46. Keller, D.; Holderegger, R. Damselflies use different movement strategies for short- and long-distance dispersal. Insect Conserv. Div. 2013, 6, 590–597. [Google Scholar] [CrossRef]
  47. Bried, J.T.; Hinchliffe, R.P. Improving Taxonomic resolution in large-scale freshwater biodiversity monitoring: An example using wetlands and Odonata. Insect Conserv. Div. 2019, 12, 9–17. [Google Scholar] [CrossRef]
  48. Šigatová, H.; Dolný, A.; Samways, M.J.; Hardersen, S.; Oliveira-Junior, J.M.B.; June, L.; Van Dinh, K.; Bried, J.T. Odonata as indicators of pollution, habitat quality, and landscape disturbance. In Dragonflies & Damselflies: Model Organisms for Ecological and Evolutionary Research, 2nd ed.; Córdoba-Aguilar,, A., Beatty, C.D., Bried, J.T., Eds.; Oxford University Press: Oxford, UK, 2023; pp. 371–399. [Google Scholar]
  49. Medina-Espinoza, E.F.; June, L.; Calvão, L.B.; Cruz, G.A. Variations in the Odonata assemblages: How do the dry season and water bodies influence them? Neotrop. Entomol. 2024, 53, 630–640. [Google Scholar] [CrossRef] [PubMed]
  50. Raebel, E.M.; Merckx, T.; Riordan, P.; Macdonald, D.W.; Thompson, D.J. The dragonfly delusion: Why it is essential to sample exuviae to avoid biased surveys. J. Insect Conserv. 2010, 14, 523–533. [Google Scholar] [CrossRef]
  51. Bried, J.T.; DAmico, F.; Samways, M.J. A critique of the dragonfly delusion hypothesis: Why sampling exuviae does not avoid bias. Insect Conserv. Div. 2012, 5, 398–402. [Google Scholar] [CrossRef]
  52. Patten, M.A.; Bried, J.T.; Smith-Patten, B.D. Survey data matter: Predicted niche of adult vs breeding Odonata. Freshw. Sci. 2015, 34, 1114–1122. [Google Scholar] [CrossRef]
  53. Kutcher, T.E.; Bried, J.T. Adult Odonata conservatism as an indicator of freshwater wetland condition. Ecol. Indic. 2014, 38, 31–39. [Google Scholar] [CrossRef]
  54. De, K.; Dey, D.; Shruti, M.; Uniyal, V.P.; Adhikari, B.S.; Johnson, J.A.; Hussain, S.A. β-diversity of odonate community of the Ganga River: Partitioning and insights from local and species contribution. Wetlands Ecol. Manag. 2023, 31, 899–912. [Google Scholar] [CrossRef]
  55. Tennessen, K.J. Dragonfly Nymphs of North America: An Identification Guide; Springer: Cham, Switzerland, 2019. [Google Scholar]
  56. Silva, L.F.R.; Castro, D.M.P.; June, L.; Callisto, M.; Hughes, R.M.; Hermes, M.G. A matter of suborder: Are Zygoptera and Anisoptera larvae influenced by riparian vegetation in Neotropical savanna streams? Hydrobiologia 2021, 848, 4433–4443. [Google Scholar] [CrossRef]
  57. Posit Team. RStudio: Integrated Development Environment for R; Posit Software; PBC: Boston, MA, USA, 2023; Available online: http://www.posit.co/ (accessed on 23 July 2024).
  58. Oksanen, J.; Blanchet, F.G.; Friendly, M.; Kindt, R.; Legendre, P.; McGlinn, D.; Minchin, P.R.; OHara, R.B.; Simpson, G.L.; Solvmos, P.; et al. Package Vegan, R Package Version 2.6-4. 2022. Available online: https://CRAN.R-project.org/package=vegan (accessed on 23 July 2024).
  59. Maechler, M.; Rousseeuw, P.; Struyf, A.; Hubert, M.; Hornik, K. cluster: Cluster Analysis Basics and Extensions, R Package Version 2.1.4. 2022. Available online: https://CRAN.R-project.org/package=cluster (accessed on 23 July 2024).
  60. Altman, N.; Krzywinski, M. Clustering. Nat. Methods 2017, 14, 545–546. [Google Scholar] [CrossRef]
  61. Legendre, P.; Gallagher, E.D. Ecologically meaningful transformations for ordination of species data. Oecologia 2001, 129, 271–280. [Google Scholar] [CrossRef] [PubMed]
  62. Poos, M.S.; Jackson, D.A. Addressing the removal of rare species in multivariate bioassessments: The impact of methodological choices. Ecol. Indic. 2012, 18, 82–90. [Google Scholar] [CrossRef]
  63. Wellborn, G.A.; Skelly, D.K.; Werner, E.E. Mechanisms creating community structure across a freshwater habitat gradient. Ann. Rev. Ecol. Syst. 1996, 27, 337–363. [Google Scholar] [CrossRef]
  64. City of Lubbock. City of Lubbock, Texas Master Drainage Plan: 2010 Update. Available online: https://ci.lubbock.tx.us/storage/images/sqXjFEEizzaq7V49ynM1VuXzWgRCxDavguVrt8Zv.pdf (accessed on 4 February 2024).
  65. Beisel, J.-N.; Usseglio-Polatera, P.; Thomas, S.; Moreteau, J.-C. Stream community structure in relation to spatial variation: The influence of mesohabitat characteristics. Hydrobiologia 1998, 389, 73–88. [Google Scholar] [CrossRef]
  66. Villalobos-Jiménez, G.; Dunn, A.M.; Hassall, C. Dragonflies and damselflies (Odonata) in urban ecosystems: A review. Eur. J. Entomol. 2016, 113, 217–232. [Google Scholar] [CrossRef]
  67. Husband, D.M.; McIntyre, N.E. Urban areas create refugia for odonates in a semi-arid region. Insects 2021, 12, 431. [Google Scholar] [CrossRef] [PubMed]
  68. Goertzen, D.; Suhling, F. Promoting dragonfly diversity in cities: Major determinants and implications for urban pond design. J. Insect Conserv. 2013, 17, 399–409. [Google Scholar] [CrossRef]
  69. Larsen, R.R. Notes on the fragile habitat, distribution, and ecology of the bleached skimmer (Libellula composita). Argia 2008, 20, 19–20. [Google Scholar]
  70. Swanson, G.A.; Winter, T.C.; Adomaitis, V.A.; LaBaugh, J.W. Chemical Characteristics of Prairie Lakes in South-Central North Dakota–Their Potential for Influencing Use by Fish and Wildlife; U.S. Fish and Wildlife Service, Fish and Wildlife Technical Report; US Department of the Interior, Fish and Wildlife Service: Washington, DC, USA, 1988; Volume 18, p. 44.
  71. Halse, S.A.; Shiel, R.J.; Williams, W.D. Aquatic invertebrates of Lake Gregory, Northwestern Australia, in relation to salinity and ionic composition. Hydrobiologia 1998, 381, 15–29. [Google Scholar] [CrossRef]
  72. Carver, S.; Storey, A.; Spafford, H.; Lynas, J.; Chandler, L.; Weinstein, P. Salinity as a driver of aquatic invertebrate colonization behavior and distribution in the wheatbelt of Western Australia. Hydrobiologia 2009, 617, 75–90. [Google Scholar] [CrossRef]
  73. Preston, T.M.; Borgreen, M.J.; Ray, A.M. Effects of brine contamination from energy development on wetland macroinvertebrate community structure in the Prairie Pothole Region. Environ. Pollut. 2018, 239, 722–732. [Google Scholar] [CrossRef] [PubMed]
  74. Hudak, P.F. Sulfate and chloride concentrations in Texas aquifers. Environ. Int. 2000, 26, 55–61. [Google Scholar] [CrossRef] [PubMed]
  75. Mengel, K. Iron availability in plant tissues—Iron chlorosis on calcareous soils. Plant Soil 1994, 165, 275–283. [Google Scholar] [CrossRef]
  76. Natural Resources Conservation Service. Soils of Texas, Texas Almanac. 2021. Available online: https://www.texasalmanac.com/articles/soils-of-texas (accessed on 4 February 2024).
  77. Dolný, A.; Pyszko, P.; Šigutová, H. Community changes in odonate monitoring: Why are long-term studies so relevant? Insect Conserv. Div. 2021, 14, 597–608. [Google Scholar] [CrossRef]
  78. Smith, L.M.; Haukos, D.A.; McMurry, S.T.; LaGrange, T.; Willis, D. Ecosystem services provided by playas in the High Plains: Potential for USDA conservation programs. Ecol. Applic. 2011, 21, S82–S92. [Google Scholar] [CrossRef]
  79. Belden, J.B.; Hanson, B.R.; McMurry, S.T.; Smith, L.M.; Haukos, D.A. Assessment of the effects of farming and conservation programs on pesticide deposition in High Plains wetlands. Environ. Sci. Technol. 2012, 46, 3424–3432. [Google Scholar] [CrossRef] [PubMed]
  80. Rosen, D.J.; Caskey, A.D.; Conway, W.C.; Haukos, D.A. Vascular flora of saline lakes in the Southern High Plains of Texas and eastern New Mexico. J. Bot. Res. Inst. Tex. 2013, 7, 595–602. [Google Scholar]
  81. Harabiš, F.; Dolný, A. Human altered ecosystems: Suitable habitats as well as ecological traps for dragonflies (Odonata): The matter of scale. J. Insect Conserv. 2012, 16, 121–130. [Google Scholar] [CrossRef]
  82. Ball-Damerow, J.E.; M’Gonigle, L.K.; Resh, V.H. Changes in occurrence, richness, and biological traits of dragonflies and damselflies (Odonata) in California and Nevada over the past century. Biodiv. Conserv. 2014, 23, 2107–2126. [Google Scholar] [CrossRef]
  83. Rocha-Ortega, M.; Rodríguez, P.; Córdoba-Aguilar, A. Spatial and temporal effects of land use change as potential drivers of odonate community composition but not species richness. Biodiv. Conserv. 2018, 28, 451–466. [Google Scholar] [CrossRef]
  84. Suárez-Tovar, C.M.; Castillo-Pérez, E.U.; Sandoval-García, I.A.; Schondube, J.E.; Cano-Santana, Z.; Córdoba-Aguilar, A. Resilient dragons: Exploring Odonata communities in an urbanization gradient. Ecol. Indic. 2022, 141, 109134. [Google Scholar] [CrossRef]
  85. Catling, P.M. A potential for the use of dragonfly (Odonata) diversity as bioindicators of the efficiency of sewage lagoons. Can. Field-Nat. 2005, 119, 233–236. [Google Scholar] [CrossRef]
  86. Barbosa de Oliveira-Junior, J.M.; De Marco Junior, P.; Dias-Silva, K.; Leitão, R.P.; Leal, C.G.; Pompeu, P.S.; Gardner, T.A.; Hughes, R.M.; June, L. Effects of human disturbance and riparian conditions on Odonata (Insecta) assemblages in eastern Amazon Basin streams. Limnologica 2017, 66, 31–39. [Google Scholar] [CrossRef]
  87. Thompson, A.L. The Life History of Anax junius (Drury) in Minnesota: Determining Instars, Growth Development Pathways, Emergence Phenology, and the Effect of Temperature on Development. Ph.D. Thesis, University of Minnesota, Minneapolis, MN, USA, December 2019. [Google Scholar]
  88. Paulson, D. Damselflies and Dragonflies of the West; Princeton University Press: Princeton, NJ, USA, 2009. [Google Scholar]
  89. Abbott, J.C. Dragonflies of Texas: A Field Guide; University of Texas Press: Austin, TX, USA, 2015. [Google Scholar]
  90. Carrillo-Lara, D.E.; Novelo-Guitérrez, R. The life cycle of Orthemis ferruginea (Fabricius, 1775) (Odonata: Libellulidae). Int. J. Odonatol. 2021, 24, 275–299. [Google Scholar] [CrossRef] [PubMed]
  91. Wissinger, S.A. Effects of food availability on larval development and inter-instar predation among larvae of Libellula lydia and Libellula luctuosa (Odonata: Anisoptera). Can. J. Zool. 1988, 66, 543–549. [Google Scholar] [CrossRef]
  92. Smith, S.G.F.; Smith, D.L. Salinity tolerance of Erythemis simplicicollis Say (Odontata: Anisoptera, Libellulidae). In Proceedings of the Sixth Symposium on the Natural History of the Bahamas, San Salvador, The Bahamas, 9–13 June 1996; pp. 139–142. [Google Scholar]
  93. Hilsenhoff, W.L. Aquatic Insects of Wisconsin: Keys to Wisconsin Genera and Notes on Biology, Habitat, Distribution and Species; Publication# G3648; University of Wisconsin, Natural History Museums Council: Madison, WI, USA, 1995; Volume 3, pp. 1–77. [Google Scholar]
  94. Hedlund, J.S.U.; Lv, H.; Lehmann, P.; Hu, G.; Anderson, R.C.; Chapman, J.W. Unraveling the world’s longest non-stop migration: The Indian Ocean crossing of the globe skimmer dragonfly. Front. Ecol. Evol. 2021, 9, 698128. [Google Scholar] [CrossRef]
  95. Catling, P.M. Climate warming as an explanation for the recent northward range extension of two dragonflies, Pachydiplax longipennis and Perithemis tenera, into the Ottawa Valley, Eastern Ontario. Ott. Field-Nat. Club 2016, 130, 122–132. [Google Scholar] [CrossRef]
  96. Zia, A.; Din, A.U.; Azam, I.; Munir, A.; Afsheen, S. Effect of salinity gradients on species composition of Odonata naiads. Arthropods 2018, 7, 11–25. [Google Scholar]
  97. Schilling, E.G.; Lawrenz, R.; Kundel, H. An assessment of the geographic distribution and status of a rare dragonfly, Rhionaeschna mutata, at the northwestern edge of its range. Northeast. Nat. 2019, 26, 523–536. [Google Scholar] [CrossRef]
  98. Bick, G.H.; Bick, J.C. Demography and behavior of the damselfly, Argia apicalis (Say), (Odonata: Coenagriidae). Ecology 1965, 46, 461–472. [Google Scholar] [CrossRef]
  99. Tennessen, K.J. Reproductive Behavior and Isolation of Two Sympatric Coenagrionid Damselflies in Florida. Ph.D. Thesis, University of Florida, Gainesville, FL, USA, 1975. [Google Scholar]
  100. Córdoba-Aguilar, A.; Rocha-Ortega, M. Damselfly (Odonata: Calopterygidae) population decline in an urbanizing watershed. J. Insect Sci. 2019, 19, 30. [Google Scholar] [CrossRef] [PubMed]
  101. Norling, U. Growth, winter preparations and timing of emergence in temperate zone Odonata: Control by a succession of larval response patterns. Int. J. Odonatol. 2021, 24, 1–36. [Google Scholar] [CrossRef] [PubMed]
  102. Runck, C.; Blinn, D.W. Secondary production of Telebasis salva (Odonata) in a thermally constant aquatic ecosystem. J. N. Am. Benthol. Soc. 1993, 12, 136–147. [Google Scholar] [CrossRef]
Diversity 16 00468 g001
Figure 1. Aerial map (Google Earth 7.3 Texas Panhandle 33°32′32″ N, 102°09′33″ W, elevation 1005 m, https://earth.google.com/web/@33.42437931,-102.0005347,1026.76427193a,96344.70435334d,35y,0h,0t,0r/data=OgMKATA (accessed 7 May 2024)) and photos of some of the wetlands of the Southern Great Plains: (a) playa, (b) urban playa lake, (c) salina, (d) salina without an active spring. Notice the variation in vegetation present at each wetland. The grey and white area at top centre of the aerial map is the city of Lubbock, Texas. The large wetlands at lower centre are salinas. Irrigated agriculture is evident as circular areas (centre-pivot irrigation). Fallow fields are tan. Grazed and ungrazed grasslands are greenish-grey.
Figure 1. Aerial map (Google Earth 7.3 Texas Panhandle 33°32′32″ N, 102°09′33″ W, elevation 1005 m, https://earth.google.com/web/@33.42437931,-102.0005347,1026.76427193a,96344.70435334d,35y,0h,0t,0r/data=OgMKATA (accessed 7 May 2024)) and photos of some of the wetlands of the Southern Great Plains: (a) playa, (b) urban playa lake, (c) salina, (d) salina without an active spring. Notice the variation in vegetation present at each wetland. The grey and white area at top centre of the aerial map is the city of Lubbock, Texas. The large wetlands at lower centre are salinas. Irrigated agriculture is evident as circular areas (centre-pivot irrigation). Fallow fields are tan. Grazed and ungrazed grasslands are greenish-grey.
Diversity 16 00468 g001
Diversity 16 00468 g002
Figure 2. Locations of the wetlands we surveyed in Texas. Symbols oversized for visualisation.
Figure 2. Locations of the wetlands we surveyed in Texas. Symbols oversized for visualisation.
Diversity 16 00468 g002
Diversity 16 00468 g003
Figure 3. Species accumulation curves of species observed at five-minute intervals at each of two locations separated by at least 50 m at four of our most speciose wetlands. Cluster analysis indicated that the wetlands we surveyed fell into three non-overlapping groups, with 9–22 wetlands per cluster (Figure 4). The wetland clusters parsed out based on environment but not according to grouping terms such as “playas”, “salinas”, or “urban playa lakes” (see summary descriptions in Table 2). Thus, although such grouping terms may be useful in indicating some similarities in physical characteristics, they do not do an adequate job of indicating just how variable the wetlands of the Southern High Plains are. Two of the clusters were primarily comprised of long-hydroperiod wetland that differed in water chemistry, particularly with respect to salinity, pH, phosphate, sulphate, and nitrate (Table 2). These findings were corroborated by PCA, which determined that at least three components were needed to explain at least 50% of the variation in the wetland data; the first three principal components explained 64.5% of the variation in the data (Figures S1 and S2). PC1 was strongly positively associated with vegetation, phosphate, and turbidity and negatively with salinity, nitrite, and sulphate. PC2 was strongly positively associated with turbidity and negatively with vegetation and phosphate. Finally, PC3 was associated positively with nitrate, sulphate, and phosphate, and negatively with turbidity and salinity (Figures S3–S5).
Figure 3. Species accumulation curves of species observed at five-minute intervals at each of two locations separated by at least 50 m at four of our most speciose wetlands. Cluster analysis indicated that the wetlands we surveyed fell into three non-overlapping groups, with 9–22 wetlands per cluster (Figure 4). The wetland clusters parsed out based on environment but not according to grouping terms such as “playas”, “salinas”, or “urban playa lakes” (see summary descriptions in Table 2). Thus, although such grouping terms may be useful in indicating some similarities in physical characteristics, they do not do an adequate job of indicating just how variable the wetlands of the Southern High Plains are. Two of the clusters were primarily comprised of long-hydroperiod wetland that differed in water chemistry, particularly with respect to salinity, pH, phosphate, sulphate, and nitrate (Table 2). These findings were corroborated by PCA, which determined that at least three components were needed to explain at least 50% of the variation in the wetland data; the first three principal components explained 64.5% of the variation in the data (Figures S1 and S2). PC1 was strongly positively associated with vegetation, phosphate, and turbidity and negatively with salinity, nitrite, and sulphate. PC2 was strongly positively associated with turbidity and negatively with vegetation and phosphate. Finally, PC3 was associated positively with nitrate, sulphate, and phosphate, and negatively with turbidity and salinity (Figures S3–S5).
Diversity 16 00468 g003
Diversity 16 00468 g004
Figure 4. Results of k-means cluster analysis to identify the optimal number of non-overlapping wetland clusters, based on environmental variables. Results for k = 2–5 are shown. Wetlands grouped into three non-overlapping clusters (k = 3).
Figure 4. Results of k-means cluster analysis to identify the optimal number of non-overlapping wetland clusters, based on environmental variables. Results for k = 2–5 are shown. Wetlands grouped into three non-overlapping clusters (k = 3).
Diversity 16 00468 g004
Diversity 16 00468 g005
Figure 5. Venn diagram of species by wetland clusters. Four-letter species codes: ANJU = Anax junius, BRGR = Brachymesia gravida, CEPO = Celithemis eponina, DYFU = Dythemis fugax, ERSI = Erythemis simplicicollis, LICP = Libellula composita, LICM = Libellula comanche, LIPU = Libellula pulchella, LISA = Libellula saturata, ORFE = Orthemis ferruginea, PALO = Pachydiplax longipennis, PAFL = Pantala flavescens, PAHY = Pantala hymenaea, PETE = Perithemis tenera, PHMI = Phanogomphus militaris, PLLY = Plathemis lydia, PLSU = Plathemis subornata, RHMU = Rhionaeschna multicolor, SYCO = Sympetrum corruptum, TESA = Telebasis salva, TRLA = Tramea lacerata, TRON = Tramea onusta, ARAL = Argia alberta, ARIM = Argia immunda, ARSE = Argia sedula, ENCI = Enallagma civile, HEAM = Hetaerina americana, ISBA = Ischnura barberi, ISDA = Ischnura damula, ISDE = Ischnura denticollis, ISRA = Ischnura ramburii, ISVE = Ischnura verticalis, LEAL = Lestes alacer, LEAU = Lestes australis.
Figure 5. Venn diagram of species by wetland clusters. Four-letter species codes: ANJU = Anax junius, BRGR = Brachymesia gravida, CEPO = Celithemis eponina, DYFU = Dythemis fugax, ERSI = Erythemis simplicicollis, LICP = Libellula composita, LICM = Libellula comanche, LIPU = Libellula pulchella, LISA = Libellula saturata, ORFE = Orthemis ferruginea, PALO = Pachydiplax longipennis, PAFL = Pantala flavescens, PAHY = Pantala hymenaea, PETE = Perithemis tenera, PHMI = Phanogomphus militaris, PLLY = Plathemis lydia, PLSU = Plathemis subornata, RHMU = Rhionaeschna multicolor, SYCO = Sympetrum corruptum, TESA = Telebasis salva, TRLA = Tramea lacerata, TRON = Tramea onusta, ARAL = Argia alberta, ARIM = Argia immunda, ARSE = Argia sedula, ENCI = Enallagma civile, HEAM = Hetaerina americana, ISBA = Ischnura barberi, ISDA = Ischnura damula, ISDE = Ischnura denticollis, ISRA = Ischnura ramburii, ISVE = Ischnura verticalis, LEAL = Lestes alacer, LEAU = Lestes australis.
Diversity 16 00468 g005
Diversity 16 00468 g006
Figure 6. Redundancy analysis results with respect to wetland hydroperiod and the two statistically significant numeric variables, categorised into relatively low, medium, and high values via Jenks Natural Breaks (for vegetation, “bare” = low, “some” = medium, “vegetated” = high). Each dot represents a wetland’s odonate assemblage.
Figure 6. Redundancy analysis results with respect to wetland hydroperiod and the two statistically significant numeric variables, categorised into relatively low, medium, and high values via Jenks Natural Breaks (for vegetation, “bare” = low, “some” = medium, “vegetated” = high). Each dot represents a wetland’s odonate assemblage.
Diversity 16 00468 g006
Table 1. Odonate species richness by site; “--“ indicates a wetland that never held water during our sampling visits. To protect landowner privacy, spatial coordinates provided only upon request from the corresponding author. “*” indicates the wetland was sampled in both 2020 and 2021. “~” indicates the wetland was sampled only in 2021.
Table 1. Odonate species richness by site; “--“ indicates a wetland that never held water during our sampling visits. To protect landowner privacy, spatial coordinates provided only upon request from the corresponding author. “*” indicates the wetland was sampled in both 2020 and 2021. “~” indicates the wetland was sampled only in 2021.
Site NameSite CodeSpecies Richness
Abernathy Community Park, Abernathy *U112
Buster Long Park, Lubbock, *U25
Clifford H. Andrews Park, Lubbock *U37
Caudle Lake Park, Hale Center *U49
K.N. Clapp Park, Lubbock *U516
George W. Dupree Park, Lubbock *U67
Edgar & Essie Givens Park, Plainview *U78
Earl Crow Park, Lubbock *U86
Charles A. Guy Park, Lubbock *U910
Frank Higinbotham Park, Lubbock *U106
Phil Hoel Park, Lubbock *U113
Jack Stevens Park, Lubbock *U124
Jan Jennings Park, Lubbock *U138
Leftwich Park, Lubbock *U149
Leroy Elmore Park, Lubbock *U152
Lobo Lark Park, Levelland *U169
George Mahon Park, Lubbock *U173
Maxey Park, Lubbock *U184
McAlister Park, Lubbock *U195
McCullough Park, Lubbock *U209
Muleshoe City Park, Muleshoe *U2110
O.W. Ribble Park, Lubbock *U224
Brookdale Remington Park, Lubbock *U233
Travis Trussel Park, Plainview *U248
Southeast Park, Amarillo ~U255
Southeast Park, Canyon ~U265
Tahoka Lake, Wilson *S130
Bean, Gaines Co. ~F29
McKenzie Lake, Gaines Co. *S220
Mound Lark, Terry Co. ~F17
Coyote Lake, Bailey Co.Coyote--
Baileyboro Lake, Bailey Co.Baileyboro--
Goose Lake, Muleshoe National Wildlife Refuge ~S47
Paul’s Lake, Muleshoe National Wildlife Refuge ~F310
Rich Lake, Terry Co. ~S316
Shafter Lake, Andrews Co.Shafter--
White Lake, Muleshoe National Wildlife Refuge ~F49
B1, Floyd Co. ~P14
B2, Floyd Co. ~P25
Castro1, Castro Co. ~P33
Cotton Center, Hale Co. ~P48
D2C, Dawson Co. ~D2C--
DarrylB, Castro Co. ~P52
Dport, Floyd Co. ~Dport--
Floyd1, Floyd Co. ~P61
Floyd2, Floyd Co. ~P72
Mil, Floyd Co. ~Mil--
Mur, Floyd Co. ~P82
TTU Range Barn, Lubbock ~P92
S26G, Swisher Co. ~S26G--
Floyd Co. 1 ~P104
Floyd Co. 2 ~P116
Floyd Co. 3 ~P127
SPart, Floyd Co. ~SPart--
SW2007_3C, Swisher Co. ~3C--
SW2007_4C, Swisher Co. ~4C--
Wson, Floyd Co. ~Wson--
Wten, Floyd Co. ~P135
Table 2. Average water chemistry and vegetation values by wetland (Site Code), with summary description of each wetland cluster. Hp = hydroperiod, categorised as relatively long or short; NO3 = nitrate (mg/L NO3); SO2 = sulphate (mg/L); PO4 = phosphate (mg/L PO43-); NO2 = nitrite (mg/L NO2-N); NTU = turbidity (nephelometric turbidity units); Veg = % vegetation cover in littoral zone; Salinity = saltiness (ppm); pH = acidity/alkalinity; Macrophytes = presence/absence (Y/N) of floating macrophytes in the littoral zone. Vegetation values (as percents) were arcsine square root-transformed; other numerical variables were Log10-transformed. Jenks Natural Breaks was used to visualise relatively low (white boxes), medium (light grey boxes), and high (dark grey boxes with white font) values of vegetation and water chemistry. Cluster 1: sites F2, F3, F4, S3, P10, U1, U3, U4, U10, U11, U13, U16, U19, U20, U21, U23. Cluster 2: F1, S1, S2, S4, P4, P6, P7, P9, U2, U6, U7, U8, U9, U12, U14, U15, U17, U18, U22, U24, U25, U26. Cluster 3: P1, P2, P3, P5, P8, P11, P12, P13, U5.
Table 2. Average water chemistry and vegetation values by wetland (Site Code), with summary description of each wetland cluster. Hp = hydroperiod, categorised as relatively long or short; NO3 = nitrate (mg/L NO3); SO2 = sulphate (mg/L); PO4 = phosphate (mg/L PO43-); NO2 = nitrite (mg/L NO2-N); NTU = turbidity (nephelometric turbidity units); Veg = % vegetation cover in littoral zone; Salinity = saltiness (ppm); pH = acidity/alkalinity; Macrophytes = presence/absence (Y/N) of floating macrophytes in the littoral zone. Vegetation values (as percents) were arcsine square root-transformed; other numerical variables were Log10-transformed. Jenks Natural Breaks was used to visualise relatively low (white boxes), medium (light grey boxes), and high (dark grey boxes with white font) values of vegetation and water chemistry. Cluster 1: sites F2, F3, F4, S3, P10, U1, U3, U4, U10, U11, U13, U16, U19, U20, U21, U23. Cluster 2: F1, S1, S2, S4, P4, P6, P7, P9, U2, U6, U7, U8, U9, U12, U14, U15, U17, U18, U22, U24, U25, U26. Cluster 3: P1, P2, P3, P5, P8, P11, P12, P13, U5.
ClusterSite CodeHpNO3SO2PO4NO2NTUVegSalinitypHMacrophytesSummary
1F2Short−0.22181.8451−1.3979−20.53910.46362.99960.9227YMostly long-
F3Short01.8451−0.9208−1.15490.47860.10022.99960.9299Yhydroperiod
F4Short0.16141.6434−0.4750−1.25960.604202.82610.9199Nwetlands of
P10Short0.25530.69900.3979−2.22180.55871.10712.16731.1464Nmoderate
S3Long−0.39791.8451−0.1938−2.22181.40821.57082.99960.8998Yturbidity,
U1Long−0.52291.2430−0.7959−1.78251.59000.65901.94960.9212Nsalinity, and
U3Long−0.52291.1139−1.0969−1.82391.67350.27742.11560.9106Nacidity, with
U4Long−0.69901.1461−0.5528−1.67781.48360.32182.25160.9268Nrelatively little
U10Long−0.52291.1761−0.5607−2.69901.61700.15882.03540.9248Nvegetation or
U11Long−0.18711.20410.1383−1.86971.583802.05230.9281Nmacrophytes.
U13Long0.23041.0212−0.8861−2.22192.14720.15882.02880.9219NModerate-to-
U16Long0.09691.1903−0.6383−1.79591.34200.07082.02880.9348Nhigh sulphate,
U19Long−0.52290.6021−1.3468−2.52291.52500.36142.00320.9292Ngenerally low
U20Long−0.25961.53150.1688−2.52291.415002.33740.9012Yphosphate and
U21Long−1.25961.84510.1222−2.30101.21460.52202.69280.9395Ynitrate, variable
U23Long−0.52291.6128−0.1427−2.12491.63600.10022.03440.9261Nin nitrite.
2F1Short−0.52291.84520.1399−1.95861.14610.32180.79100.8932YMostly long
P4Short0.98681.44720.3979−1.82391.06821.57081.94600.8751Nhydroperiods,
P6Short1.05310.60210.3979−1.88612.66181.57082.16440.9375Nmoderate-to-
P7Short0.94451.84510.3979−1.92081.35601.57081.98410.8774Nhigh nitrate,
P9Short0.62320.69900.3979−1.39792.13351.57082.11060.8669Nsulphate,
S1Long−0.11541.8451−0.7280−2.27302.25861.12830.49090.8850Ynitrite,
S2Long0.55021.84510.1658−1.82391.83340.78540.77340.8834Yphosphate, and
S4Long0.23041.84510.3424−2.69900.75050.937700.8739Nturbidity. Low-
U2Long0.26711.32220.1761−2.18711.60800.32182.26780.9261Nto-moderate
U6Long0.09691.39790.1903−1.54521.84140.39772.14920.8822Nsalinity and
U7Long0.19030.81290.3284−2.45591.66560.65902.05480.8618Yacidity.
U8Long0.14610.69900.3979−2.39791.49480.15881.96520.8859NVariable with
U9Long1.19730.97780.1688−2.69902.69940.15882.24540.9513Yrespect to
U12Long−0.22181.27880.1320−2.02201.59600.15882.00060.8993Nvegetation
U14Long0.16140.81290.2467−2.34680.94370.22552.07370.9031Nand
U15Long0.38021.64340.1123−2.09691.31280.15882.04980.9101Nmacrophytes.
U17Long0.20411.32220.1446−0.59351.62321.20941.82830.8727N
U18Long−0.02230.88620.9164−1.73281.23040.22551.88310.9028N
U22Long−0.02231.51190.3979−1.90311.73920.10022.06600.9149N
U24Long0.27881.38920.1303−1.81361.42080.63302.00300.9199Y
U25Long0.30101.74040.3979−2.69900.97681.10712.53910.8675N
U26Long−0.09691.5315−0.0809−2.30101.35410.93772.08640.8976N
3P1Short−0.30100.30100.3979−2.69900.72751.57082.30960.8621NMostly short
P2Short−0.52280.47710.3979−1.95861.62531.57082.26720.8445Nhydroperiod
P3Short−0.52280.3010−0.1308−2.699031.57082.07190.8825Nwith shoreline
P5Short−0.52280.30100.3979−30.75661.57082.19310.8432Nvegetation but
P8Short−0.52280.3010−0.0555−2.69902.17610.93771.99210.9222Nfew floating
P11Short0.14610.95420.3979−2.39791.04141.17312.01280.8561Nmacrophytes.
P12Short−0.22180.30100.4116−2.69900.60421.04722.11730.8887YGenerally low
P13Short−0.52280.30100.1173−2.69902.82221.24901.91860.8859Nin nitrate,
U5Long0.14610.78030.2480−2.60210.57340.99122.14320.8420Ysulphate, nitrite,
and acidity.
Generally high
in phosphate,
moderate
salinity.
Variable in
turbidity.
Table 3. Site by species (presence-absence) table by wetland cluster. Odonates are listed in alphabetical order starting with dragonflies from the left, followed by damselflies. Light grey cells indicate species found at only a single wetland; dark grey cells with white font indicate species found at multiple wetlands but within a single cluster. Four-letter species codes are as in Figure 5. Cluster 1: sites F2, F3, F4, S3, P10, U1, U3, U4, U10, U11, U13, U16, U19, U20, U21, U23. Cluster 2: F1, S1, S2, S4, P4, P6, P7, P9, U2, U6, U7, U8, U9, U12, U14, U15, U17, U18, U22, U24, U25, U26. Cluster 3: P1, P2, P3, P5, P8, P11, P12, P13, U5.
Table 3. Site by species (presence-absence) table by wetland cluster. Odonates are listed in alphabetical order starting with dragonflies from the left, followed by damselflies. Light grey cells indicate species found at only a single wetland; dark grey cells with white font indicate species found at multiple wetlands but within a single cluster. Four-letter species codes are as in Figure 5. Cluster 1: sites F2, F3, F4, S3, P10, U1, U3, U4, U10, U11, U13, U16, U19, U20, U21, U23. Cluster 2: F1, S1, S2, S4, P4, P6, P7, P9, U2, U6, U7, U8, U9, U12, U14, U15, U17, U18, U22, U24, U25, U26. Cluster 3: P1, P2, P3, P5, P8, P11, P12, P13, U5.
SITECLUSTERANJUBRGRCEPODYFUERSILICPLICMLILULIPULISAORFEPALOPAFLPAHYPETEPHMIPLLYPLSURHMUSYCOTESATRLATRONARALARIMARSEENCIISBAISDAISDEISDNISHAISRAISVELEALLEAUHEAM
F211000000000000110100101100011000000000
F311000000010110110000101100010000000010
F411000000001000110000101100010000000010
S311000100011110110000111100111001000000
U111000000000111111100100100010000000000
U310000000000001111000001100010000000000
U410000000000110111000101000010000000001
U1111000000000000001000000000010000000000
U1311000000000000110000101100010000100000
U1610000000000110101000101100011000000000
U1911000000000000110000100000010000000000
U2011000000000011101000100100011000000000
U2111000100011100100001101000010000000000
U2310000000000000010000000100010000000000
P1011000000000000110000000000010000000000
F121000000000000010000101010011000000000
P421000000001000100000101100010000000010
P620000000000000000000000000010000000000
P720000000000000010000000000010000000000
P920000000000000000000100000010000000000
S121101111111111101111101111011111001101
S221000100001111111101101111011000000100
S421000000000000110000101100010000000000
U220000000000000100000101000011000000000
U620000000000001101000101100010000000000
U721000000001000110000101100010000000000
U821000000001000001000100100010000000000
U921000000000011101000101100011000000000
U1021000000000000101000100100010000000000
U1220000000000010000000100000011000000000
U1421000000001010110000101100010000000000
U1520000000000000000000100000010000000000
U1721000000000000100000000000010000000000
U1820000000000000100000100000010000000000
U2220000000000000010000100100010000000000
U2420000100001001010001100100010000000000
U2521000000010001100000000000010000000000
U2620000100000000000000000000011000011000
P131000000000000000000101000010000000000
P231000000000000100000101000010000000000
P330000000000000100000001000010000000000
P530000000000000000000000000010000000010
P830000000000000100000000000010000000000
P1131000000000000110000101000010000000000
P1231000000000000110000101100010000000000
P1331000000000000110000001000010000000000
U531010100011001111101111100010000000000
Table 4. Odonate richness by wetland cluster.
Table 4. Odonate richness by wetland cluster.
ClusterNumber of WetlandsNumber of Odonate Species
11623
22232
3917
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

Husband, D.M.; McIntyre, N.E. Finding Isolated Aquatic Habitat: Can Beggars Be Choosers? Diversity 2024, 16, 468. https://doi.org/10.3390/d16080468

AMA Style

Husband DM, McIntyre NE. Finding Isolated Aquatic Habitat: Can Beggars Be Choosers? Diversity. 2024; 16(8):468. https://doi.org/10.3390/d16080468

Chicago/Turabian Style

Husband, Danielle M., and Nancy E. McIntyre. 2024. "Finding Isolated Aquatic Habitat: Can Beggars Be Choosers?" Diversity 16, no. 8: 468. https://doi.org/10.3390/d16080468

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