Invasion of Egeria densa Planch.

in
semiarid reservoirs

Julia Brito Lacet, Ariadne do

Nascimento Moura & Ênio Wocyli
Dantas

Brazilian Journal of Botany

ISSN 0100-8404

Braz. J. Bot
DOI 10.1007/s40415-019-00548-9

1 23
Your article is protected by copyright and
all rights are held exclusively by Botanical
Society of Sao Paulo. This e-offprint is for
personal use only and shall not be self-
archived in electronic repositories. If you wish
to self-archive your article, please use the
accepted manuscript version for posting on
your own website. You may further deposit
the accepted manuscript version in any
repository, provided it is only made publicly
available 12 months after official publication
or later and provided acknowledgement is
given to the original source of publication
and a link is inserted to the published article
on Springer's website. The link must be
accompanied by the following text: "The final
publication is available at link.springer.com”.

1 23
 Author's personal copy
Brazilian Journal of Botany
https://doi.org/10.1007/s40415-019-00548-9

ORIGINAL ARTICLE

Invasion of Egeria densa Planch. in semiarid reservoirs

Julia Brito Lacet1 · Ariadne do Nascimento Moura2 · Ênio Wocyli Dantas3

Received: 6 September 2018 / Revised: 5 August 2019 / Accepted: 16 August 2019

© Botanical Society of Sao Paulo 2019

Abstract
The aim of the present study was to relate the vegetation cover of Egeria densa to climatic, limnological and diversity vari-
ables to gain an understanding of how they affect occurrence and colonization of the macrophyte. This study was conducted in
six reservoirs of the Paraíba do Norte River Basin. Six sampling units were established in each reservoir, and the occurrence
of vegetation cover of the aquatic plant was verified at three-month intervals. The occurrence was evaluated based on the
occurrence of E. densa, and the establishment phase was evaluated based on vegetation cover. The macrophyte was found
in three reservoirs and exhibited invasive behavior in two of them. PERMANOVA and logistic regression analyses were
performed to understand the occurrence of the species. Precipitation, total dissolved solids and richness were evidenced as
the significant variables. PERMANOVA and generalized linear models were used to evaluate the colonization, indicating
that precipitation and temperature were determinant factors of the invasion of E. densa. The model generated for occurrence
was robust and had a 4.3-fold higher explanatory power than the model of colonization. Results revealed that occurrence
is the most important aspect for the successful invasion of E. densa and, once present in an environment, its colonization
is probable.

Keywords Aquatic macrophytes · Colonization · Invasive species · Occurrence · Plant cover

1 Introduction

Biodiversity is one of the most valued environmental

resources due to its benefits to humanity and the essential
services of ecosystems (Rai 2015). Thus, the dissemination
of invasive species constitutes a threat to biodiversity.
Biological invasion is not a simple process of the entrance
* Ênio Wocyli Dantas of an exotic species in a new environment. For an invasive
eniowocyli@yahoo.com.br species to establish itself, the effective colonization of the
Julia Brito Lacet environment is necessary, characterized by the reproduction
julacet@hotmail.com and survival of the taxon (Williamson 2006). Generally, the
Ariadne do Nascimento Moura success of the invasion is related to the characteristics of
ariadne_moura@hotmail.com the population of origin (Ribas et al. 2018). Invasions are
1
Programa de Pós‑graduação em Botânica, Universidade
determined by biological and environmental factors (Theo-
Federal Rural de Pernambuco-UFRPE, Rua D. Manoel de harides and Dukes 2007), such as the evolutionary history of
Medeiros, s/n, Dois Irmãos, Recife, PE CEP 52171‑900, the environment, community structure, propagule pressure,
Brazil environmental disturbances, stress and phenotypic plasticity
2
Departamento de Biologia, Botânica, Universidade Federal (Alpert et al. 2000; Riis et al. 2010). As an invasion may not
Rural de Pernambuco-UFRPE, Rua D. Manoel de Medeiros, be successful, understanding the factors that influence the
s/n, Dois Irmãos, Recife, PE CEP 52171‑900, Brazil
occurrence and colonization of the species in the environ-
3
Centro de Ciências Biológicas e Sociais Aplicadas‑CCBSA, ment can assist in the development of management strate-
Departamento de Ciências Biológicas, Campus V,
Universidade Estadual da Paraíba-UEPB, R. Horácio Trajano
gies for the prevention of further invasions and the effective
de Oliveira, s/n, Cristo, João Pessoa, PB CEP 58070‑450,
Brazil

13
Vol.:(0123456789)
 Author's personal copy
J. B. Lacet et al.

control of those that have already taken place (Azzurro et al. 2016). Two climatic types are found in the Paraíba do Norte
2006; Horst et al. 2014). River Basin: one downstream (As) and one upstream (Bsh)
Some species are known to achieve considerable inva- (Alvares et al. 2013). This study was developed in reservoirs
sion success in environments where they are introduced. of the Bsh (hot semiarid) climate region with temperatures
An example of this is the macrophyte Egeria densa Planch, ranging from 18 to 31 °C (Azevêdo et al. 2015). The annual
which is native to Argentina, Uruguay and southern Brazil, rainfall averages are less than 800 mm; average sunshine is
but has been widely distributed in temperate zones through- 2800 h year−1, evaporation is 2000 mm year−1 and relative
out the world due to the aquarium trade (Durand et al. 2016). humidity is about 50%. Therefore, the region has a negative
E. densa is listed as an invasive species by the International water balance (Sabino et al. 2015).
Union for Conservation of Nature due to its multiple impacts
in aquatic environments (Gillard et al. 2017). Sampling – Sampling was performed at three-month inter-
This plant forms large monospecific banks with no vals between March 2015 and July 2016. Rainfall was
genetic diversity (Thiébaut et al. 2016). Reproduction by scarcer in this period than normal (mean annual precipita-
seeds is rare, since plants of the same sex are commonly tion less than 250 mm) because the region was under the
found living in the same environment (Mori et al. 2012). effect of the climate phenomenon El Niño. Six reservoirs
E. densa develops well in polluted locations contaminated were studied (Boqueirão, Camalau, Cordeiro, Mucutu,
with heavy metals (Mori et al. 2012), has a C4 metabolism Namorados and Poções) (Table 1). Six marginal plots
(Casati et al. 2002), grows in both lotic and lentic environ- measuring 2500 m2 were established at each reservoir as
ments (Yarrow et al. 2009) and survives in a temperature the sampling units. The exception was Namorados, in which
range of 10º to 25 °C, with optimum growth at a temperature only two sampling units were established due to its small
of 16 °C (Thiébaut et al. 2016). The macrophyte is found volume. A sampling unit was determined on each bank in
in cold climates in places like Alaska, Canada, the USA, the different regions of the reservoirs: river, lake and dam.
Europe and Japan (Fominykha et al. 2016). In tropical and Subsampling was performed in the sampling units through
subtropical regions, it is limited to high altitudes or sources the demarcation of three transects perpendicular to the bank,
of cold water (Thiébaut et al. 2016). Despite its characteris- along which quadrants (0.5 X 0.5 m) were established, with
tics linked to cooler climates, E. densa has been recorded in a distance of two meters between quadrants. Throughout
the semiarid region of northeast Brazil (Oliveira et al. 2005; the study, 160 sampling units were evaluated, but 32 were
Santos et al. 2006; Bezerra et al. 2007; Nascimento et al. excluded for being dry at the time of sampling. Thus, a total
2008; Pereira et al. 2008; Moro et al. 2014; Azevêdo et al. of 128 effective sampling units were considered in the pre-
2015; Sabino et al. 2015). sent study.
Considering the expansion of E. densa to the semiarid
region of Brazil and discussions on problems stemming from In all sampling units, we collected data on richness, per-
biological invasions, the aim of the present study was to centage of cover of each macrophyte species and abiotic
determine possible factors that enabled the occurrence and variables. Richness was defined as the total number of taxa
the colonization of the aquatic plant in the semiarid region. present in each sampling unit. All taxa in the plots were
collected for the identification of the species and deter-
mination of richness. The identification of the plants was
2 Materials and methods performed based on the specialized literature and consulta-
tions with specialists. The classification system followed
Study area – The hydrographic basin of the Paraíba do the rules of APG IV (APG IV 2016).
Norte River has a total area of 20,071.83 km2 and located in The quantification of the vegetation cover was per-
northeast Brazil between latitudes 6°51′31’’S and 8°26′21’’S formed using the Braun-Blanquet (1979) method. For
and longitudes 34°48′35’’W and 37°2′15”W (Costa et al.

Table 1  Maximum Reservoirs Location River Area ­(105 ­m2) Volume ­(106 ­m3)
morphometric data of the
reservoirs studied in the state of Boqueirão (BOQ) 07°28′S e 36°08′O Paraíba 325.96 411.68
Paraiba, Brazil
Camalau (CAM) 07°53′S e 36°50′O Paraíba 22.34 48.11
Cordeiro (COR) 07°49′S e 36°41′O Paraíba 52.18 69.97
Mucutu (MUC) 07°07′S e 36°39′O Taperoá 16.92 25.37
Namorado (NAM) 07°23′S e 36°32′O Taperoá 1.40 2.12
Poções (POÇ) 7°53′S e 37°01′O Paraíba 45.30 29.86

13
 Author's personal copy
Invasion of Egeria densa Planch. in semiarid reservoirs

submerged plants, the PVI correction factor was applied 3 Results

according to Canfield et al. (1984).
Stiers et al. (2011) determined three categories of inva- Description of the community – We identified 25 species
sion based on percentage of cover: (A) plots in a nonin- of aquatic plants distributed among 19 genera and ten fami-
vaded water body (0%); (B) plots in a semi-invaded water lies. Maximum total richness was found in Namorados and
body (< 25%); and (C) strongly invaded plots (> 75%). For Camalaú, with eight species in each reservoir. However,
the present study, we adapted this method considering that Boqueirão and Camalaú had the highest average richness per
sampling units with E. densa cover < 25% corresponded to sampling unit. Most species occurred in only one reservoir.
the introduction phase and sampling units with E. densa The exceptions were E. densa, Echinochloa sp., Ipomoea
cover > 25% indicated the development of the community sp., Najas marina L., Olyra sp., and Potamogeton illinoensis
and therefore corresponded to the establishment phase. Morong.
Limnological data (water temperature, pH, conductivity,
turbidity, dissolved oxygen and total dissolved solids) were Occurrence and cover of Egeria densa – Egeria densa was
measured using a HORIBA © U-50 multiparameter probe found in three reservoirs (Boqueirão, Camalau and Cord-
in each sampling unit. Water transparency was determined eiro), occurring in 62 (39.24%) of the 158 sampling units.
using a Secchi disk. Water samples were collected for the In Cordeiro, the taxon did not exhibit invading behavior and
determination of the concentrations of total nitrogen, total was found in only three sampling units (0.02% cover). In
phosphorus, nitrite, nitrate, ammonia and orthophosphate in Boqueirão, the macrophyte was absent from only one sam-
the laboratory according to APHA (1992). Monthly precipi- pling unit throughout the study. The lowest of cover was
tation at each reservoir was obtained from the Paraíba State 0.25% and the highest was 80.67%. No invasive behavior
Water Management Agency at the data collection station was found in most sampling units (63.33%). In Camalau,
closest to each reservoir. E. densa occurred in all sampling units, with cover ranging
from 0.02% to 65.51%, but exhibiting invasive potential in
Data analysis – Multivariate Analysis of Permutations only three sampling units (Table 2).
(PERMANOVA) was used to understand the occurrence of
the species. For such, the limnological, climatic and rich- Factors that influence the occurrence of E. densa – Sampling
ness variables were evaluated, considering the presence/ units with E. densa had significantly different limnologi-
absence of the taxon as a factor. The Shapiro test was used cal variables compared to environments without the species
to determine the normality of the data. We then applied the (F = 22.46, R2 = 0.0971, p < 0.001). The sampling units in
Wilcoxon test to determine which variables were different which the taxon occurred had warmer waters, greater trans-
in this factor. parency, higher concentrations of dissolved oxygen and
lower salinity. Sampling units without E. densa were more
Multiple logistic regression was used to analyze the rela- turbid, had higher total dissolved solids and conductivity as
tionship between the matrix of the environmental factors well as higher levels of nitrogen and phosphorus in the water
and the presence/absence of E. densa in all sampling units column (Table 3).
in order to generate an explanatory model (that with the
lowest AIC value) that determined which variables favored The logistic regression test had an AIC of 50.802, which
the occurrence of the species. was considered good for evaluating the reliability of the
We used a PERMANOVA to evaluate whether the per- model. Precipitation, total dissolved solids and richness
centage of E. densa differed in the sampling units in terms were selected as predictors of the occurrence of the spe-
of the limnological, climatic and richness variables, con- cies. According to the model, the occurrence of E. densa
sidering the sampling units with the highest percentage of was favored by the richness of the community, whereas
cover and those with less than 25.00% as a factor. The data precipitation and total dissolved solids were negatively
were then tested for normality using the Shapiro test, and the associated with its occurrence (Table 4).
Wilcoxon test was applied to define which variables differed
significantly between occurrence and colonization. Factors that influence the colonization of E. densa – PER-
A generalized linear model (GLM) of the “Gaussian” MANOVA found no significant differences in the limnologi-
family was used to understand what determines the coloni- cal variables according to the percentage of E. densa cover
zation, and we selected the model with the lowest AIC value. (F = 0.3568, R2 = 0.0059, p > 0.05). However, the Wilcoxon
For such, the vegetation cover vector was used only for the test demonstrated significant differences in transparency and
sampling units with E. densa and the matrix of limnological temperature between sampling units with and without the
variables. All statistical analyses were performed using the colonization of the species (Table 3).
R Program 3.3.1.

13
 Author's personal copy
J. B. Lacet et al.

Table 2  Average values (maximum) of plant cover per reservoir in the state of Paraiba, Brazil
Taxa BOQ CAM COR MUC NAM POÇ

Characeae
Chara braunii C.C. Gmelin – – 0.22 (5.17) – – –
Chara rusbyana M. Howe 0.02(0.33) – – – – –
Nitella sp. – – – – 0.38 (2.56) –
Nymphaeaceae
Nymphaea pulchella DC. – – – – 0.17 (2.15) –
Nymphaea rudgeana G. Mey. – – – – 0.01 (0.08) –
Araceae
Lemna sp. – – – – – 0.04 (1.32)
Pistia stratiotes L. – – – – – 0.04 (1.3)
Hydrocharitaceae
Egeria densa Planch. 25.30(80.67) 16.91 (65.51) 0.02 (0.02) – –
Najas arguta Kunth – 0.06 (0.80) – – – –
Najas marina L. 0.68(18.32) – 0.004 (5.73) 1.68 (11.4) 0.39 (4.63) –
Juncaginaceae
Potamogeton illinoensis Morong – 0.001 (0.02) – 1.10 (7.09) – –
Pontederiaceae
Eichhornia crassipes (Mart.) Solms – 0.15 (4.43) – – – –
Poaceae
Echinochloa sp. – – – 0.02 (0.02) 0.003 (0.04) 1.59 (23.57)
Eriochloa sp. – – 0.19 (6.32) – – –
Hymenachne sp. – – – – 0.003 (0.04) –
Olyra sp. – 0.12 (1.56) – 0.53 (4.75) 1.25 (10.61) 0.29 (8.07)
Streptochaeta sp. – 0.02 (0.02) – – – –
Catapodium sp. – – – – – 0.66 (6.01)
Tni sp. – – – – 1.83 (13.02) –
Trachypogon sp. 0.51(3.2) – – – – –
Ceratophyllaceae
Ceratophyllum demersum L. 1.01(7.29) – – – – –
Polygonaceae
Polygonum acuminatum Kunth – – – – – 0.11 (3.39)
Polygonum ferrugineum Weed. – 0.04 (1.11) – – – –
Convolvulaceae
Ipomoea sp. – 0.002 (0.04) – – – 0.02 (0.04)
Average richness per SU 2.17 1.67 0.68 0.93 1.17 0.83
Richness per Reservoir 5 8 4 4 8 7

BOQ Boqueirão, CAM Camalau, COR Cordeiro, MUC Mucutu, NAM Namorados, POÇ Poções, SU Sample Unit

The colonization model had a much higher AIC (549.95) model, selecting richness (biotic variable), total dissolved
than the introduction model. Therefore, the model cannot be solids (abiotic variable) and precipitation (climatic variable)
considered reliable, although it selected explanatory vari- as significant to the presence of the E. densa.
ables (temperature and precipitation) (Table 4). Richness favored the occurrence of E. densa. This aquatic
plant was normally present in environments with other sub-
merged macrophytes. This process of facilitation has been
4 Discussion described, and an initial invasion can facilitate a secondary
invasion, as a first invading species can cause changes in
In the present study, we considered variables that influ- environmental conditions that are similar to the needs of a
enced the models generated as favorable or unfavorable to second invader (Thiébaut and Martinez 2015). Some other
the occurrence and colonization. Occurrence had a robust submerged species in the environments with E. densa in the

13
 Author's personal copy
Invasion of Egeria densa Planch. in semiarid reservoirs

Table 3  Average values ± standard deviation of abiotic variables according to the introduction and establishment of Egeria densa in reservoirs in
the state of Paraíba, Brazil
Parameters Introduction of E. densa Establishment of E. densa
Absence of E. densa Presence of E. densa Test Cover < 25% Cover Test
>25%
n = 96 n = 62 n = 49 n = 13

Transparency (m) 1.77 ± 1.78 3.94 ± 3.23 W = 343 4.14 ± 3.47 3.17 ± 1.99 W = 4356
Precipitation (mm) 88.87 ± 81.73 66.49 ± 54.27 W = 2574. 67.48 ± 54.40 62.75 ± 55.83 W = 294
Water temperature (°C) 24.24 ± 1.78 24.89 ± 1.18 W = 3878 24.73 ± 1.21 25.48 ± 0.86 W = 437.5
pH 9.40 ± 0.74 9.58 ± 0.86 W = 3450 9.60 ± 0.78 9.69 ± 1.02 W = 391.5
Conductivity (uS/cm) 18.78 ± 30.53 9.04 ± 22.15 W = 1296 9.7 ± 23.41 6.57 ± 17.11 W = 367.5
Turbidity (NTU) 205.83 ± 217.87 88.2 ± 78.02 W = 1846 87.84 ± 83.05 89.57 ± 57.87 W = 367.5
Dissolved oxygen (mg/L) 8.06 ± 12.87 8.20 ± 9.94 W = 3889 6.40 ± 1.87 6.50 ± 2.12 W = 340.5
Total dissolved solids (mg/L) 4.82 ± 3.39 1.14 ± 0.55 W = 1198 1.14 ± 0.62 1.14 ± 0.21 W = 406.5
Salinity (ppt) 0.97 ± 1.79 0.13 ± 0.13 W = 1469 0.14 ± 0.13 0.12 ± 0.11 W = 378
Ammonium ion (μg L−1) 277.13 ± 904.18 142.69 ± 174.53 W = 2640 140.73 ± 176.59 150.1 ± 173.31 W = 318.5
Nitrous oxide (μg L−1) 13.25 ± 45.19 5.78 ± 9.22 W = 3044 6.26 ± 9.50 3.98 ± 8.18 W = 275.5
Nitrate (μg L−1) 38.01 ± 60.43 23.88 ± 14.82 W = 3380 24.43 ± 15.26 21.81 ± 13.42 W = 289.5
Total nitrogen (μg L−1) 306.67 ± 160.65 146.15 ± 93.73 W = 1259 151.38 ± 96.36 126.42 ± 83.57 W = 279
Orthophosphate (μg L−1) 121.74 ± 177.23 31.81 ± 25.14 W = 1934 30.93 ± 23.94 35.15 ± 30.09 W = 333
Total phosphorus (μg L−1) 332.49 ± 379.41 114.72 ± 222.65 W = 840 123.73 ± 249.49 80.75 ± 37.95 W = 351

W Wilcoxon test value, n number of sample units in each category. Significant values in bold

Table 4  Results of the logistic regression for introduction of E. densa such as erosion, are common, particularly in arid regions
and the GLM for the establishment of E. densa. Key: AIC = Akaike (Tanner et al. 1993). This increases the dispersion of light,
Information Criteria, R2 = determination coefficient
altering the euphotic zone and limiting the depth of sub-
Models Parameters Coefficients p value merged macrophytes due to the negative impact on photo-
synthetic rates (Kosten et al. 2011). Thus, for E. densa to be
Introduction of E. Intercept −4.14 0.001
densa introduced in semiarid environments, variables with a posi-
AIC = 50.802 Precipitation −0.02 0.04 tive influence need to be stronger than those with a negative
R2 = 0.56 Total dissolved solids −2.61 0.002 influence. As the macrophyte occurred in environments with
Richness 1.57 0.01 other submerged macrophytes, these other plants had modi-
Establishment of E. Intercept −207.61 0.005 fied the environment such that E. densa could overcome the
densa negative effect of total dissolved solids.
AIC = 549.95 Precipitation −0.1 0.055 Another limiting factor was precipitation, which is a
R2 = 0.13 Temperature 9.27 0.002 primary source of advection and directly affects the biota
by modifying the redox potential of the water, its organic
content and nutrient exchanges (Wetzel and Likens 2000).
present study are also invaders, such as Najas marina and Precipitation could have had positive relationship with
Ceratophyllum demersum. Moreover, the canopy formed by the occurrence of the species by increasing the fragmen-
different aquatic plants enables the anchorage of fragments tation rate (Redekop et al. 2016). However, the study was
(Durand et al. 2016), which are produced by self-fragmen- conducted during the period of El Niño, which drastically
tation, self-induced abscission, the activity of aquatic birds reduced rainfall in the region. Therefore, we believe that
and fishes as well as anthropogenic actions (Redekop et al. the rare rains were not sufficient to exert this positive effect.
2016). We believe that these ideas may explain the positive Moreover, precipitation can also favor the occurrence of
value attributed to richness in the occurrence model of E. submerged plants, as it leads to an increase in suspended
densa. solids, which impedes the penetration of light into the water
Total dissolved solids were the second variable selected (Silva et al. 2008). Due to this advection effect, precipitation
by the model, which indicated a negative association with and total dissolved solids are closely related, since climatic
the occurrence of the E. densa. This result was expected, as conditions can influence erosive processes, especially in the
levels of abiotic turbidity derived from natural processes, semiarid region. This region has a crystalline foundation

13
 Author's personal copy
J. B. Lacet et al.

near the surface that hinders the infiltration of water fol- REHISA—Rede Hidrográfica do Semiárido, coordinated by José
lowing rainfall, which results in more intense surface runoff Etham de Lucena Barbosa (Grant No. 1107/13). We also thank the
support of CNPQ—Conselho Nacional de Desenvolvimento Científico
(Medeiros et al. 2015) and favors advection processes. e Tecnológico, and the Postgraduate Program in Botany of UFRPE—
Colonization produced a weak explanatory model, which Universidade Federal Rural de Pernambuco.
selected precipitation and temperature as the variables cor-
related with the coverage of E. densa in the semiarid region. Author contributions JBL was responsible for collecting, processing
In the study by Oliveira et al. (2005), E. densa took approxi- and analyzing data, as well as writing the text. ANM worked on data
analysis and revision of the text. EWD participated in the project plan-
mately three months to develop in the presence of favora- ning, data analysis and review of the text.
ble substrate, nutrient, luminosity, temperature and current
speed conditions. Light was the most important factor favor-
ing the growth of the macrophyte, which develops more eas-
ily in shallower areas, although it can grow in deeper loca- References
tions due to its efficient photosynthetic uptake system. The
authors also found that a lack of light due to diverse aspects Alpert P, Bone E, Holzapel C (2000) Invasiveness, invasibility and the
of turbidity limited the colonization and growth of the spe- role of environmental stress in the spread of non-native plants.
cies, which is in agreement with the present findings, as Perspect Plant Ecol 3:52–66
Alvares CA, Stape JL, Sentelhas PC, Gonçalves JLM, Sparovek G
precipitation had a negative effect for the occurrence and the (2013) Köppen’s climate classification map for Brazil. Meteorol
colonization of the E. densa. The same effects that hindered Z 22:711–728. https​://doi.org/10.1127/0941-2948/2013/0507
the occurrence of the aquatic plant are also believed to have APG IV (2016) An update of the Angiosperm Phylogeny Group clas-
hindered the growth of its populations. However, as rainfall sification for the orders and families of flowering plants: APG IV.
Bot J Linn Soc 181:1–20
levels were low, due partially to the El Niño phenomenon, APHA, Awwa, WPCF (1992) Standard methods for the examination of
this barrier was not sufficient to impede the colonization of water and waste-water. APHA/AWWA/WPCF, New York
E. densa in a number of sampling units. Azevêdo DJS, Barbosa JEL, Porto DE, Gomes WIA, Molozzi J (2015)
Temperature was the other variable selected in the Diversity measures in macroinvertebrate and zooplankton commu-
nities related to the trophic status of subtropical reservoirs: con-
model and is known to influence the growth of the E. densa. tradictory or complementary responses? Ecol Indic 50:135–149.
According to Gillard et al. (2017), temperature alters the https​://doi.org/10.1590/S2179​-975X2​914
metabolism of plants, changes the responses of dormancy Azzurro E, Golani D, Bucciarelli G, Bernardi G (2006) Genetics of the
and growth and even favors the occurrence of invasive spe- early stages of invasion of the Lessepsian rabbitfish Siganus luri-
dus. J Exp Mar Biol Ecol 333:190–201. https​://doi.org/10.1016/j.
cies. Yarrow et al. (2009) found that temperature was the jembe​.2005.12.002
most critical factor impacting the metabolism of the mac- Bezerra TP, Silva CP, Lopes JP (2007) Utilização da macrófita aquática
rophyte. Tanner et al. (1993) found that the growth of E. Egeria densa Planchon, 1849 (Hydrocharitacea) na produção de
densa was accentuated by high temperatures counteracting tijolos para construção civil. Rev Bras Eng Pesca 2:114–127
Braun-Blanquet J (1979) Fitossociologia: bases para el estúdio de las
the adverse effects of turbidity caused by rainfall. As high comunidades vegetales. Blume, Madrid
temperatures are common in the semiarid region of northeast Canfield DE, Shireman JV Jr, Colle DE, Haller WT, Watkins IICE,
Brazil, the development of submerged species is enhanced Maceina MJ (1984) Prediction of chlorophylla concentrations
in warm waters, overcoming any negative effects of other in Florida lakes: importance of aquatic macrophytes. Can J Fish
Aquat Sci 41:497–501
variables. Casati P, Lara MV, Andreo CS (2002) Regulation of enzymes involved
In summary, understanding the occurrence of E. densa in in C4 photosynthesis and the antioxidant metabolism by UV-B
the semiarid region is more important than its colonization, radiation in Egeria densa, a submersed aquatic species. Photo-
since the model generated to explain the establishment was synth Res 71:251–264
Costa DF, Barbosa JEL, Dantas EW (2016) Productivity-diversity rela-
weak and the model of occurrence had an explanatory power tionships in reservoir phytoplankton communities in the semi-arid
4.3 times greater. It is possible that the fragility of the colo- region of northeastern Brazil. J Arid Environ 129:64–70. https​://
nization model is related to the lack of significant differences doi.org/10.1016/j.jarid​env.2016.02.010
between the environments with different percentages of cov- Durand J, Fleenor W, McElreath R, Santos MJ, Moyle P (2016) Physi-
cal Controls on the Distribution of the Submersed Aquatic Weed
erage of E. densa and the small variation in the selected vari- Egeria densa in the Sacramento-San Joaquin Delta and Implica-
ables. Thus, based on the findings of the present study, the tions for Habitat Restoration. San Franc Estuary Watershed Sci
occurrence of E. densa is the most important aspect of the 14:2–20. https​://doi.org/10.15447​/sfews​.2016v​14iss​1art4​
invasion of the species. Once found, probably the plant will Fominykh AS, Mukhutdinov VF, Kipriyanov LM (2016) Findings of
Brazilian Elodea in Cooling Ponds of the Verkhnii Tagil Power
colonize the environment, because colonization is influenced Plant (Middle Urals). Russ J Biol Invasions 7:189–194
by faint variations in precipitation and temperature. Gillard M, Grewell BJ, Deleu C, Thiébaut G (2017) Climate warm-
ing and water primroses: germination responses of populations
Acknowledgements Funding for this work was provided by from two invaded ranges. Aquat Bot 136:155–163. https​://doi.
FINEP—Financiadora de Estudos e Projetos, through the Project org/10.1016/j.aquab​ot.2016.10.001

13
 Author's personal copy
Invasion of Egeria densa Planch. in semiarid reservoirs

Horst JL, Kimball S, Becerra JX, Noge K, Venable DL (2014) Docu- differences caused by phenotypic plasticity or local adaptation?
menting the early stages of invasion of Matthiola parviflora and Ann Bot 106:813–822. https​://doi.org/10.1093/aob/mcq17​6
predicting its spread in North America. Southwest Nat 59:47–55. Sabino JHF, Araújo ES, Cotarelli VM, Siqueira Filho JA, Campelo
https​://doi.org/10.1894/F07-FMO-13.1 MJA (2015) Riqueza, composição florística, estrutura e formas
Kosten S, Jeppesen E, Huszar VLM, Mazzeo V, Nes E, Peeters ETHM, biológicas de macrófitas aquáticas em reservatórios do semiárido
Scheffer M (2011) Ambiguous climate impacts on competition nordestino, Brasil. Nat Online 13:184–194
between submerged macrophytes and phytoplankton in shal- Santos E, Gomes SO, Lopes JP (2006) Contribuição de elódea Egeria
low lakes. Freshw Biol 56:1540–1553. https​://doi.org/10.111 Densa à piscicultura através da colonização do camarão-canela
1/j.1365-2427.2011.02593​.x Macrobrachium Amazonicum no Sub-Médio Rio São Francisco,
Medeiros PRP, Cavalcante Segundo GH, Magalhães EMM (2015) no nordeste do Brasil. Rev Bras Eng Pesca 1:102–118
Comportamento da turbidez e material em suspensão, em um rio Silva AEP, Angelis CF, Machado LAT, Waichaman AV (2008) Influên-
com vazão regularizada por sistema de barragens em cascata: Rio cia da precipitação na qualidade da água do Rio Purus. Acta Ama-
São Francisco (NE, Brasil). Geochim Bras 29:35–44. https​://doi. zon 38:733–742
org/10.5327/Z0102​-98002​01500​01000​4 Stiers I, Crohain N, Josens G, Triest L (2011) Impact of three aquatic
Mori ES, Martins D, Velini ED, Marino CL, Gouvêa CF, Leite SMM, invasive species on native plants and macroinvertebrates in
Camacho E, Guries RP (2012) Genetic diversity in Egeria temperate ponds. Biol Invasions 13:2715–2726. https​: //doi.
densa and E. najas in Jupiá Reservoir, Brazil. Cien Inv Agric org/10.1007/s1053​0-011-9942-9
39:321–330 Tanner CC, Clayton JS, Wells RDS (1993) Effects of suspended sol-
Moro MF, Sousa DJL, Matias LQ (2014) Rarefaction, richness estima- ids on the establishment and growth of Egeria densa. Aquat Bot
tion and extrapolation methods in the evaluation of unseen plant 45:299–310
diversity in aquatic ecosystems. Aquat Bot 117:44–51. https:​ //doi. Theoharides KA, Dukes JS (2007) Plant invasion across space and
org/10.1016/j.aquab​ot.2014.04.006 time: factors affecting nonindigenous species success during four
Nascimento PRF, Pereira SMB, Sampaio EVSB (2008) Biomassa stages of invasion. New Phytol 176:256–273. https​://doi.org/10.1
de Egeria Densa nos Reservatórios da Hidroelétrica de Paulo 111/j.1469-8137.2007.02207​.x
Afonso-Bahia. Planta Daninha 26:481–486 Thiébaut G, Martinez L (2015) An exotic macrophyte bed may facili-
Oliveira NMB, Sampaio EVSB, Pereira SMB, Moura-Júnior AM tate the anchorage of exotic propagules during the first stage of
(2005) Capacidade de regeneração de Egeria densa nos res- invasion. Hydrobiologia 746:183–196. https​://doi.org/10.1007/
ervatórios de Paulo Afonso, BA. Planta Daninha 23:363–369 s1075​0-014-1982-6
Pereira SMB, Nascimento PRF, Sampaio EVSB, Carvalho MFO, Thiébaut G, Gillard M, Deleu C (2016) Growth, regeneration and colo-
Moura-Júnior AM (2008) Monitoramento e manejo da macrófita nization of Egeria densa fragments: the effect of autumn tempera-
aquática Egeria densa Planchon no nordeste brasileiro. Estudo de ture increases. Aquat Ecol 50:175–185. https​://doi.org/10.1007/
Caso. In: Moura AN, Araújo EL, Albuquerque UP (eds) Biodi- s1045​2-016-9566-3
versidade, Potencial econômico e Processos Eco-fisiológicos em Wetzel RG, Likens GE (2000) Limnological analyses. Springer, New
Ecossistemas Nordestinos. Nupeea, Recife, pp 209–234 York
Rai PK (2015) Concept of plant invasion Ecology as Prime Factor for Williamson M (2006) Explaining and predicting the success of invad-
Biodiversity Crisis: introductory review. Int Res J Environ Sci ing species at different stages of invasion. Biol Invasions 8:1561–
4:85–90 1568. https​://doi.org/10.1007/s1053​0-005-5849-7
Redekop P, Hofstra D, Hussner A (2016) Elodea canadensis shows Yarrow M, Marín VH, Finlayson M, Tironi A, Delgado LE, Fischer
a higher dispersal capacity via fragmentation than Egeria densa F (2009) The ecology of Egeria densa Planchon (Liliopsida:
and Lagarosiphon major. Aquat Bot 130:45–49. https​://doi. Alismatales): A wetland ecosystem engineer? Rev Chil Hist Nat
org/10.1016/j.aquab​ot.2016.01.004 82:299–313
Ribas LGs, Cássia-Silva C, Petsch DK, Silveira MJ, Lima-Ribeiro
MS (2018) The potential invasiveness of an aquatic macrophyte Publisher’s Note Springer Nature remains neutral with regard to
reflects founder effects from native niche. Biol Invasions 20:3347– jurisdictional claims in published maps and institutional affiliations.
3355. https​://doi.org/10.1007/s1053​0-018-1780-6
Riis T, Lambertini C, Olesen B, Clayton JS, Brix H, Sorrel BK (2010)
Invasion strategies in clonal aquatic plants: are phenotypic

13