https://orcid.org/0000-0001-8319-6667
Figures
Abstract
Biological invasions increase with the intensity of globalization, human activities, and climate change. Insects represent a high potential of invasive species due to their adaptability to new environment. We analysed here the potential of an Asian phytophagous bug, Corythauma ayyari (Heteroptera, Tingidae) to become widespread, recently recorded in Europe, and that depends on Jasminum spp., an ornamental plant widespread in Europe. We modelled its current distribution, projected it into the future and tested its niche overlap between native and invaded areas. When considering the host plants as environmental variables, the analysis shows that C. ayyari shifted to a new ecological niche but its distribution is restricted by its host plant distribution. Including or excluding the host plants as environmental variables has an impact on C. ayyari distribution. We recommend to consider host plant interactions when dealing with niche modelling of phytophagous species.
Citation: Durand M, Guilbert E (2024) Corythauma ayyari (Insecta, Heteroptera, Tingidae) depends on its host plant to spread in Europe. PLoS ONE 19(3): e0295102. https://doi.org/10.1371/journal.pone.0295102
Editor: Faham Khamesipour, Kerman University of Medical Sciences, ISLAMIC REPUBLIC OF IRAN
Received: November 14, 2023; Accepted: February 21, 2024; Published: March 26, 2024
Copyright: © 2024 Durand, Guilbert. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: The author(s) received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Introduction
An invasive species is an non-native species that begins to proliferate in a new geographical habitat. As such, it usually becomes a pest. A pest is a native or non-native species disrupting the environment. The invasion is often characterized by an introduction, a naturalization and a spread [1] and its success is context dependent [2].
While the number of biological invasions continues to increase, particularly with the intensity of globalization and the acceleration of climate change, they are now one of the main causes of the current loss of biodiversity. Trade would be the major driver of alien species introductions at least for invertebrates and algae [3]. In order to fight against their proliferation and the damages, it is necessary to identify these species and to analyse their invasiveness.
Insects are the most diverse group of living species on earth, thanks in part to their ability to adapt to new environmental conditions. It is therefore not surprising that many insect species are considered invasive or potentially invasive. Insects are in fact the most disrupting class of invasive species, spreading diseases, consuming crops, disrupting forest ecosystem [4, 5]. For example, 50% of recently invasive species belong to the order Hemiptera [6]. Corythauma ayyari (Drake) belongs to the family Tingidae (Hemiptera, Heteroptera). It has been first described from India (Drake 1933) and found on Jasminum pubescens (Retz.), today synonymized with J. multiflorum (Burm.). It is notably found in Pakistan, Sri Lanka, Laos, Thailand, Malaysia, India, Indonesia, and Singapore [6–8]. It is known as a pest of Jasmine in Southern India [9, 10]. However, very few specimens were recorded from these areas, as most collecting were casual. Recent observations report its presence in France [8, 11], Italy [7, 8, 12], Spain [6], Tunisia [13], Israel [14], and Syria [15]. Most records mention the three main host plants Jasminum officinale L., J. grandiflorum L., and J. sambac L. (Oleaceae) but also Volkameria inermis L. (Lamiaceae) [15] and Trachelospermum sp. (Apocynaceae) [11].
It is hypothesized to have been accidentally introduced in Europe several times independently [8] probably with Jasminum plants widely used in public and private gardens, as they are highly valued for their fragrance [6, 8]. However, the maintenance of C. ayyari on Jasmine species is not without damage to the host plant. It has been observed that adults and nymphs develop on the underside of the leaves and feed on sap. The excreta of the tingid mar the leaves, greatly decreasing photosynthesis, which deteriorates the palisade parenchyma and leads to progressive desiccation of the plant [6–8, 12, 15]. The recent arrival in Europe of C. ayyari suggests that it could wreak havoc on Jasmine species but also other ornamental and/or crop plants such as Volkameria or Trachelospermum species, causing significant environmental and economic damage.
Species Distribution Models (SDM) are increasingly used to answer biogeographic and ecological questions. Indeed, these models allow to estimate the probability of presence of a species according to different biotic and abiotic variables, in order to expose the favorable areas for its presence [16, 17]. These tools are particularly interesting when studying invasive or potentially invasive species. Previous studies have indeed demonstrated the interest and reliability of these methods to quantify the probability of presence and niche differences of an invasive species [18, 19]. Niche differences can be easily detected when variables driving species distribution are known [18], which is particularly relevant when dealing with invasive species. Close match was found between habitat suitability in potential invaded areas and pattern of occurrences of invasive species using bioclimatic niche modelling [19].
Phytophagous insects such as Tingidae which are sap suckers, are strictly dependent of their host plant. Particularly if they are monophagous. Therefore, their distribution highly depends on the distribution of the host plant. Two publications have already focused on the current distribution of C. ayyari at the level of the Mediterranean basin and specially in the Iberian Peninsula [6] and in Italy [8]. Therefore, it makes sense to reinforce them by focusing here on the invasive potential of the species.
The aim of this study is to analyse the invasive potential of C. ayyari by modelling its current distribution and its potential ecological niche in Europe, according to the current climatic conditions and the distribution of its main host plants. We modelize C. ayyari distribution including or not the four main host plants (J. officinale, J. multiflorum, J. grandiflorum and J. sambac), as environmental variable to evaluate their influence on C. ayyari distribution. On this basis, C. ayyari potential distribution is projected into the future. The native and the invasive ecological niches are compared to see if a shift from native to invasive niche is observed.
Materials & methods
Study site
The study focuses on the potential spread area of C. ayyari, according to the distribution of its host plants and considering their native area in Asia. It includes an area between 18° West to 141° East and between 1° to 70° North; covering the distribution area of Jasminum species in Europe and the Mediterranean Basin between 18° West to 40° East and 30° to 70° North, and the native area which covers a surface within 1° to 46° North and 65° to 141° East. It covers also part of Africa, where occurrences of Jasminum species are known. If occurrences of the host plants are known from North America, none of C. ayyari is known there.
Species occurrences
Occurrences of C. ayyari were retrieved from the GBIF network (dio:10.15468/dl.b3jj26), from scientific publications [8, 11, 13, 20, 21] and also from personal communications. Many other specimens of C. ayyari are known but were collected long time ago and without accurate location as to be used in this study. Similarly, occurrences of J. officinale, J. grandiflorum, J. multiflorum, and J. sambac, were extracted from the GBIF network (dio:10.15468/dl.9sbc2q and dio:10.15468/dl.2j7zqa). Jasminum officinale has an extended distribution in Europe, compared to the three other Jasminum species. There are only 21 records of J. grandiflorum in South West Europe, the rest are located in Asia. Jasminum multiflorum is sparcely recorded from South East Asia and totalizes two records in Europe (Germany and Greece). Jasminum sambac is also widely distributed in Asia but is mainly recorded in India. There are only five records of J. sambac in North-East Europe. We did not include Trachelospermum sp. as host plant, as the species has not been identified, and Volkameria inermis as it is a tropical species, essentially distributed on the seashore in South East Asia between India and the South Pacific islands. No field survey was done in this study to acquire original data. As such no permits were necessary.
Environmental variables
We used the 19 climatic variables available in WorldClim version 2.0 (www.worldclim.org) at a spatial resolution of 2.5 minutes. We also used the potential distribution of the four Jasminum species as biotic variable for C. ayyari. The procedure is explained below.
Niche modelling
Before modelling the distribution of species, we analyse the correlation between variable using ‘Hmisc’ package and Pearson coefficient, either for the current and the future conditions. The sets of current and future climatic variables with a correlation coefficient up to 0.7 were different. As collinearity affects models trained on data at different times, we opted for variables issued from a principal component analysis (PCA) [22]. We did a PCA of the 19 climatic variables either of the current and future conditions using ‘FactoMineR’ package, and we selected the first nine PCA components as variables, accounting for 99% of the variance. Jasminum species modelled distributions were also used as biotic variables as they were not correlated up to 0.7 to the PCA components used as variables, and as Jasminum species and C. ayyari distributions were not correlated.
Data were cleaned before modelling using ‘CoordinateCleaner’ package [23], to get rid of inadequate occurrences such as the specimens deposited in museums, occurrences wrongly geo-referenced or duplicated. We considered the tests ‘urban_ref’, ‘capitals_ref’, as null since Jasminum species are ornamental plants and therefore can be present in urban environments.
Pseudo-absences were added to the dataset of occurrences. We selected the same amount of pseudo-absences as occurrences as suggested by Bardet-Massin et al. [16]. We randomly selected 10 sets of pseudo-absences for each Jasminum species, assuming that their distribution is in equilibrium since they were introduced in Europe and the Mediterranean Basin long time ago. Assuming that the distribution of C. ayyari is not in equilibrium [24] as we deal with an invasive species, we selected pseudo-absence out of a convex hull calculated on the basis of the known occurrences [25]. According to several authors, low sample size does not affect model accuracy as soon as the species have small geographic range and restricted environmental tolerance [26, 27], which seems to be the case here. In addition, each dataset analysed here was up to 20–30 observations, which is the minimum sample size required to ensure model accuracy under generic procedures of modelling as stated by Jiménez-Valverde [28].
In a first step, we modelized the potential distribution of Jasminum species considering PCA variables, and projected it into the future (2081–2100 period). In a second step, we did a first modelling of C. ayyari potential distribution, considering PCA variables only, and we did a second one considering PCA variables and including the rasters resulting of the modelled distributions of Jasminum species as biotic variables. Prior to this last analysis, we tested the correlation between C. ayyari distribution and the one of Jasminum species. The two modelling were compared to see the influence of Jasminum species distribution on C. ayyari modelling. We projected the distribution of C. ayyari into the 2081–2100 period as well, using only PCA variables corresponding to the period, and in a second step including also the modelled projection into the future of Jasminum species. For the projection into the future, we considered SSP585 scenario, the worst scenario of global warming, corresponding to an additional radiative forcing of 8.5 W/m2, and we use the CNRM-CM6-1 model, which presents particularly fine parameter calibration strategies tailored to the geographic area considered [29]. Modelling was made using ‘Biomod2’ package [30] and the following options:
All the models were tried except Maxent which is the only one based on presence only, i.e. Generalized Linear Model (GLM), Generalized Additive Model (GAM), Generalized Boosting Model (GBM), Classification Tree Analysis (CTA), Artificial Neural Network (ANN), Surface Range Envelop (SRE), Flexible Discriminant Analysis (FDA), Random Forest (RF), and Multiple Adaptive Regression Splines (MARS).
We run 10 evaluations, split the data in 20% test data and 80% modelling data. Models were selected to be combined using a threshold of 0.7 on the basis of TSS evaluation. Modelling was evaluated by ensemble forecasting, combining all model as suggested by [31], with an arbitrary threshold up to 0.7 and using the TSS binary metric which is independent to prevalence, as recommended by [32]. Combined model was evaluated using Boyce index with ‘ecospat’ package [33, 34]. Boyce index measures how much model predictions differ from random distribution of the observed distribution. It ranges between -1 and +1. A value of +1 indicates the model fits perfectly with the distribution and a value of zero indicates the model is not different from a random distribution.
Niche comparison
We also compared ecological niches of Jasmium species and C. ayyari in the invaded area. Ecological niche estimates were performed with the ‘ecospat’ package [34, 35]. A Mantel test was done first to test spatial autocorrelation between variables. Niche overlap between native and invaded areas was also measured using the Schoener’s D index [35–37]. It calculates the occupancy of the environment by the species. The comparison of this occupancy between two areas or two species is used to calculate the niche overlap. It can range from 0, for niches with no overlap, to 1, for completely identical niches. A niche equivalency and a niche similarity test are performed. The equivalency tests whether the niche overlap is constant when randomly reallocating the occurrences of two species among two ranges. The similarity tests whether the occupied niche in one range is more similar to the one occupied in the other range than expected by random. We evaluated niche overlap on the basis of the 19 climatic variables. We compare niche overlap only for C. ayyari, J. grandiflorum and J. officinale, as the two other Jasminum species do not have enough occurrences in the invaded area to be analysed.
Results
Jasminum officinale totalizes 1381 occurrences restricted to the areas studied, of which 78% are located in Europe. It totalizes 1065 occurrences after cleaning. Jasminum sambac totalizes 626 occurrences after cleaning over 724 existing. Jasminum grandiflorum totalizes 186 occurrences after cleaning over 203 existing, and J.multiflorum totalizes 27 occurrences after cleaning over 40 existing. Corythauma ayyari totalizes 48 occurrences, of which 37 are located in Europe and 11 are located in Asia. It totalizes 34 occurrences after cleaning (S1 Data) (Fig 1).
Map of the study area showing the occurrences of C. ayyari (red diamonds), and its four main host plants, J. officinale (black dots), J. grandiflorum (orange dots), J. multiflorum (green dots), and J. sambac (blue dots).
Niche modelling
The results of current distribution modelling and projection into the future of Jasminum species is illustrated in supplementary information (S1 Fig). Boyce index is high for J. officinale (0.952), J. grandiflorum (0.946) but low for J. multiflorum (0.65), and J. sambac (0.668). Including Jasminum distributions as variable, Boyce index for C. ayyari distribution modelling is 0.760; while without Jasminum variables it is 0.659.
Considering only PCA components, Corythauma ayyari current distribution modelling shows a high suitable habitat on the North Mediterranean coasts in Spain, France, Italy, the Croatian coasts, the Aegean and Marmara coasts, the South of Black Sea coast, part of North of France,Belgium and the Netherlands, at a lower degree parts of North and Central Europe except some central areas, and North Scandinavia. In Asia, suitable areas include India except West coast and North, Thailand, Myanmar, Laos, Vietnam and Cambodgia except West coast of South East Asia, Malaysia included, and a large part of South and East China (Fig 2A).
Projections of the current (upper row) and future (lower row) environmental conditions for C. ayyari using climatic variables only (left colon), and including also Jasminum species distribution as variables (right colon). The scale represents percentage of suitability along a gradient of colors.
Corythauma ayyari distribution projected into the future shows a reduction of the current suitable areas mainly Spain, Italia, the North of Mediterranean and Marmara coasts, also moving North of Europe facing the Channel and the North Sea, parts of Scandinavia where some unsuitable areas turned 50% suitable (Fig 2B). In Asia, East China and most areas North to India turned unsuitable.
Correlation between Jasminum species distribution and C. ayyari distribution are low (ranging between 0.32 to 0.46, S1 Table). Correlation between Jasminum species does not exceed 0.53 (between J. multiflorum and J. grandiflorum), lesser than the usual threshold accepted for correlated variables in niche modelling which is 0.7. As such, including Jasminum species distribution as variables does not bias modelling.
When including Jasminum species distribution as variables, C. ayyari current distribution modelling shows suitable areas reduced to North Mediterranean coasts, including Aegean and Marmara coasts. It includes also part of West Europe and Maghreb coasts with a lesser suitability. In Asia, areas are not highly suitable, however, they include Part of South and East India, central South East Asia, except Malaysia and a large region in South East China (Fig 2C).
Corythauma ayyari distribution projected into the future with Jasminum species distributions reduces suitable areas to a thin range along Mediterranean coasts, enlarging suitable areas to most of South Mediterranean coasts, Portugal, the Atlantic side of Spain and South West of France. Suitability of areas is reduced in Asia, except in West India in Gujarat region and Pakistan (Fig 2D).
The most contributing variables for C. ayyari modelling for the ensemble forecasting are the 5th, 1st and 2nd PCA components (39%, 15% and 12%, respectively). When including Jasminum distribution as variables, the three most contributing variables are the 5th PCA component (28%) and J. officinale and J. grandiflorum distribution modeling (23% and 13%, respectively) (S2 Table, Fig 3A and 3B).
PCA plot of the 19 bioclimatic variables on the first two axes, A) for the current conditions, and B) for the future conditions. The graphs show the 19 bioclimatic variables contribution on the two first axes of the PCA for the current conditions and for the future conditions (2081–2100).
Niche comparison
Overlap of Jasminum species and C. ayyari niches in the native area is low between all species (Table 1), except between J. multiflorum and J. grandiflorum (70%), for which the similarity is significant.
Overlap between species in the invaded niche is significantly similar only between J. officinale and C. ayyari. It has not been tested for J. multiflorum and J. sambac due to the too low number of occurrences in Europe (Table 2).
Comparison of native and invaded niche for C. ayyari shows a very low overlap (D = 0.024). It is neither equivalent, nor similar (P value = 0.9604 and 0.0594, respectively).
Discussion
The biology of C. ayyari is only known by one study [9] conducted in India on J. sambac. It stated that C. ayyari develops at a temperature ranging from 22 and 31°C, and a relative humidity ranging from 86.75 to 91.88%. Adults emerge at night and mate 24 hours after emergence. Females live between 9 to 12 days and mate with an average of 34 eggs.
Corythauma ayyari seems to thrive in the warm-summer Mediterranean climate of the European regions [38–40]. This temperate climate (minimum temperatures between -3°C and 18°C throughout the year) offers dry and warm summers, with precipitation less than 40 mm and maximum temperatures above 22°C, and mild winters [39]. The period of greatest activity of C. ayyari, would probably occur between late spring and early autumn, coinciding with the warm summers of the Mediterranean climate, while in winter during the coldest periods, C. ayyari would overwinter [12, 40, 41].
The native distribution of C. ayyari is restricted to tropical climates (minimum temperature greater than or equal to 18°C) of equatorial savannah with dry winter (rainfall in the driest month less than 60 mm), and to warm temperate climates with dry winter and hot summer (average temperature of the hottest month greater than 22°C) [30, 40]. The temperature and the presence of a dry period are thus important variables common to both native and invaded areas of C. ayyari. Low temperatures seem to be the main constraints to the establishment of C. ayyari. Thermophilic in nature, it survives, like most other Tingidae, at a temperature range of 15°C to about 40°C, with an optimal development and fecundity rate around 30°C [8, 10, 41–43]. At lower temperatures, it enters hibernation. The mild winter of the Mediterranean climate should therefore be favourable to it. Thus, under the worst-case scenario ssp585, temperatures will increase by 3.3°C to 5.7°C, for the period 2081–2100 [44]. This increase in temperature in the Mediterranean climate may facilitate optimal development of C. ayyari, while reducing its hibernation periods in future years.
Jasminum species are mainly recorded from their native area in Asia, except J. officinale which is mainly recorded from Europe, due to its importation as ornamental plant. Then, distribution modelling shows that Europe is mainly suitable for J. officinale, but weakly suitable for J. multiflorum, and not suitable for J. sambac. Mediterranean coasts only are suitable for J. grandiflorum.
Not surprisingly, the main suitable areas for C. ayyari are in the native region, South East Asia, despite few occurrences recorded there, comparing to the occurrences in Europe. Outside its original range, according to our models when considering climatic variables only, the most suitable areas for C. ayyari are concentrated on the North coasts of Mediterranean Sea, where it has been recently recorded, and where J. officinale and J. grandiflorum are present. Parts of North West of Europe remain also suitable for C. ayyari. Not surprisingly also, suitable areas are restricted to a thinner range along North Mediterranean coasts and part of Western Europe at a lower suitability, when Jasminum species are included as variables. This is mainly due to J. officinale and J. grandiflorum distributions, as J. sambac has no suitable areas in Europe, and J. multiflorum has a low suitability in Northern Europe. As expected, C. ayyari suitable area is restricted by its host plants distribution.
Projections into the future for J. grandiflorum and J. officinale show that suitable areas will be restricted to a narrower band along Mediterranean coasts and will also move West of Europe. In consequence, C. ayyari will also be restricted to the same areas, according to the projection including Jasminum species as variable. Corythauma ayyari suitable areas will extend North and East of Europe according to modelled distribution without Jasminum variables. That could be unconvincing, considering low temperatures in Central Europe during winter, and the near absence of Jasminum in this region. Jasminum is absent in the East of Europe, except two occurrences of J. sambac near Stockholm and near Kharkiv (East of Ukraine), and an occurrence of J. officinale near Tallinn.
Modelling of C. ayyari distribution and projection into the future differs when including or not Jasminum species as variable. Jasminum distribution modelling shows that C. ayyari will not benefit from its main host plants to spread out in Europe as would suggest modelling using only climatic variables. In addition, including the host plants as biotic variables render the modelling more accurate (see Boyce index). This show the importance of integrate biotic variables such as host plants for phytophagous species in modelling. Nevertheless, model validation is low (Boyce index around 0.7). Modelling would benefit from a larger sample size of J. sambac and C. ayyari which are around 30 occurrences each, nevertheless sufficient according to Jimenez-Valverde’s criteria [28]. Small geographic and environmental ranges can also positively affect model performance [27, 45]; which seems to be the case of host plants and C. ayyari in the non-native range.
At the present day, the suitable area of C. ayyari in Europe does not fit with J. officinale distribution and is restricted to Mediterranean coasts, like J. grandiflorum. Harbours are the main entrance of the host plants, and then C. ayyari would have been introduced in Europe with its host plants [8]. It was recorded on J. grandiflorum in Spain and Syria; J. sambac and J. grandiflorum in Tunisia; J. sambac and J. multiflorum in Israel; J. officinale and J. grandiflorum in Italia. However, the invaded niche of C. ayyari and of J. officinale are similar (P value = 0.0297), while this is not the case of J. grandiflorum’s niche, and Europe is not suitable for J. sambac. As native and invaded niche of C. ayyari is neither equivalent nor similar, this would suggest that C. ayyari shifted to one niche to another, and could potentially benefit from the presence of J. officinale to setting up in Europe. As niche modelling shows, J. officinale is one of the most contributing variables (23%). More South in Mediterranea, J. sambac and J. grandiflorum could stay the main plant hosting C. ayyari. Unfortunately, it is difficult to know the contribution of climatic variables to niche modelling as we used PCA components to avoid auto-correlation between variables. Another point to consider also is that the low overlap between Jasminum species and C. ayyari in the native area may be due to the lack of occurrences as no studies were conducted on the relationships between these species in Asia. Most of the records of C. ayyari in the native range is due to casual collecting. Overlap in Europe is mainly due to the fact that C. ayyari is recorded specifically as non-native species on cultivated or ornamental plants. However, Broenniman et al. [18] used kernel density function to determine a smoothed density of occurrences when estimating ecological niche. That makes moving from geographical space, where the species occur, to a multivariate environmental space, where analyses are performed, independent of sampling effort.
Niche shift of introduced plant species is not rare as stated by Atwater et al. [46]. This assumption highly depends on variable use and modelling method and has been debated on invasive plants [47, 48], and also on introduced mammals [18].
The adaptability of C. ayyari suggest that it could spread to other host plants. Indeed, C. ayyari has been observed on the genera Lantana, Ocimum, Musa, Hedychium, and Trachelospermum, and the species Althea officinalis, Daedalacanthus nervosus, and Volkameria inermis [7, 12–14]. The observation of C. ayyari on these plants does not mean they are host plants but may be only visited plants. Most of them are popular ornamental plants and widespread in gardens so, their trade would be the main source of the introduction of C. ayyari and one of the major factors for its expansion. For example, Trachelospermum jaminoides is probably the Trachelospermum species on which C. ayyari was found in France [11], as it is the only species of the genus recorded in Europe. It has a wide distribution in Europe and could be a potential host, facilitating C. ayyari spread.
Conclusions
This study shows that host plants distribution is key to phytophagous species. They may restrict invasive potential of the invader. However, oligo- and polyphagous species have the possibility to shift to several host plants and then invade new areas, shifting from one niche to another. Halyomorpha halys (Stal, 1855) is a typical example of a successful invasive species as it is highly polyphagous and feeds on more than 120 wild or cultivated plant species (Streito et al. 2021). Modelling distribution depends on data availability, as invasive species provide scarce data when early considered. Therefore, modelling their potential invasiveness is often done once species already established and disrupted native ecosystem.
In several countries like India, Pakistan, Egypt, Tunisia, Jasminum, especially J. grandiflorum and J. sambac, have a very high symbolic value and are emblematic plants of the country (Haouas et al. 2015). They are also used in the perfume industry. The invasion of C. ayyari in the non-native regions would cause significant economic but also symbolic damage. Therefore, as human activity is the main source of C. ayyari expansion due to the ubiquitous presence of ornamental plants in gardens and parks, means of management in the latter are conceivable, and would be different from a widely cultivated plant as food resource.
Supporting information
S1 Data. GPS coordinates of occurrences of the species used in the analyses.
GPS coordinates of Jasminum officinale (Joff), J. grandiflorum (Jgra), J. multiflorum (Jmul), J. sambac (Jsam) and C. ayyari (Cay) occurrences after cleaning (see method section), used for analyses.
https://doi.org/10.1371/journal.pone.0295102.s001
(DOCX)
S1 Fig.
Projections of the current (on the left) and future (on the right) environmental conditions resulting from the niche modelling for Jasminum species. Ensemble projection combines all model with an arbitrary threshold up to 0.7 and uses the TSS binary metric, for the four main Jasminum species hosting C. ayyari: A, B) J. officinale; C, D) J. grandiflorum; E, F) J. multiflorum; G, H) J. sambac.
https://doi.org/10.1371/journal.pone.0295102.s002
(TIF)
S1 Table. Correlations between species distribution.
Correlation between species distribution using Pearson coefficient, under current condition (above) and future conditions (below). Species distributions are rasters resulting from species distribution modelling using PCA components as variables.
https://doi.org/10.1371/journal.pone.0295102.s003
(DOCX)
S2 Table. Contribution of variables used for modelling C. ayyari distribution.
A first modelling uses only the first nine climatic variables issued from a PCA; a second modelling includes also the four Jasminum species distribution as variable.
https://doi.org/10.1371/journal.pone.0295102.s004
(DOCX)
Acknowledgments
The authors thanks Alexandre Schickele for providing help with R scripts, and Jean-Claude Streito for fruitful comments on the manuscript.
References
- 1. Richardson DM, Pysek P, Rejmanek M, Barbour MG, Panetta FD, West CJ. Naturalization and invasion of alien plants: concepts and definitions. Diversity and Distributions. 2000; 6: 93–107.
- 2. Hui C, Richardson DM, Landi P, Minoarivelo HO, Garnas J, Roy HE. Defining invasiveness and invasibility in ecological networks. Biological Invasions. 2016; 18: 971–983.
- 3. Seebens H, Blackburn T, Dyer E, et al. No saturation in the accumulation of alien species worldwide. Nature Communications. 2017; 8: 14435. pmid:28198420
- 4. Bradshaw C, Leroy B, Bellard C, Roiz D, Albert C, Fournier A et al. Massive yet grossly underestimated global costs of invasive insects. Nature Communications. 2016; 7, 12986. pmid:27698460
- 5. Venette RC., Hutchison W.D. (2021) Invasive Insect Species: Global Challenges, Strategies & Opportunities. Frontiers in Insect Science. 2021; 1. pmid:38468878
- 6. Roca-Cusachs M, Goula M. First record of the invasive tingid species Corythauma ayyari (Drake, 1933) in the Iberian Peninsula (Insecta: Hemiptera: Heteroptera: Tingidae). Bulletín de la Institució Catalana d’Història Natural. 2014; 8: 119–123.
- 7. Carapezza A. Corythauma ayyari (Drake, 1933) new pest of jasmine in Italy (Heteroptera Tingidae). Naturalista Siciliano. 2014; 38: 381–384.
- 8. Mazza G, Nerva L, Strangi A, Mori E, Chitarra W, Carapezza A, et al. Scent of Jasmine Attracts Alien Invaders and Records on Citizen Science Platforms: Multiple Introductions of the Invasive Lacebug Corythauma ayyari (Drake, 1933) (Heteroptera: Tingidae) in Italy and the Mediterranean Basin. Insects. 2020; 11: 620. pmid:32927750
- 9. Nair RCP, Nair MRG. Studies on the biology of the lace-wing Corythauma ayyari Drake, a pest of Jasmine. Agricultural Research Journal of Kerala. 1974; 12:172–173.
- 10.
Schaefer CW, Panizzi AR (Eds.). Heteroptera of economic importance. CRC press. 2000.
- 11. Streito JC, Matocq A, Guilbert E. Découverte d’un foyer de Corythauma ayyari (Drake 1933) et point sur la présence de plusieurs espèces de Stephanitis envahissants en France (Hemiptera: Tingidae). L’Entomologiste. 2010; 66: 7–12.
- 12. Pedata PA, Guilbert E, Nugnes F, Mancini D. Rinvenimento di una popolazione di Corythauma ayyari (Heteroptera Tingidae) su Jasminum officinale (Oleaceae): un nuovo fitofago per l’Italia. Ricerca. 2013; 3: 36–39.
- 13. Haouas D, Guilbert E, Ben Halima-Kamel M. First report of Corythauma ayyari (Drake) (Hemiptera: Tingidae) on Arabian and Spanish jasmine in Tunisia. EPPO Bulletin. 2015; 4:, 144–147.
- 14. Novoselsky T, Freidberg A. Note: Corythauma ayyari (Drake) (Hemiptera: Heteroptera:Tingidae) a new pest of ornamentals in Israel. Phytoparasitica. 2013; 41: 149–150.
- 15. Zeity M, Ali AY. First report of the lacebug Corythauma ayyari (Drake) (Hemiptera: Tingidae) on Jasminum grandiflorum L. and Jasminum sambac (L.) from Syria. EPPO Bulletin. 2019; 49: 398–400.
- 16. Bardet-Massin M, Jiguet F, Albert CH, Thuiller W. Selecting pseudo-absences for species distribution models: how, where and how many? Methods in Ecology and Evolution. 2012; 3: 327–338.
- 17. Bilger I, Paillet Y, Laguet S, Doutau B, Bullifon F, Chevalier M, et al. Modélisation de la distribution des petites chouettes de montagne dans les Alpes du Nord françaises. Naturae. 2021; 13: 167–181.
- 18. Broennimann O, Fitzpatrick MC, Pearman PB, Petitpierre B, Pellissier L, Yoccoz NG, et al. Measuring ecological niche overlap from occurrence and spatial environmental data. Global Ecology Biogeography. 2012; 21: 481–497.
- 19. Thuiller W, Richardson DM, Pyšek P, Midgley GF, Hughes GO, Rouget M. Niche‐based modelling as a tool for predicting the risk of alien plant invasions at a global scale. Global change biology. 2005; 11: 2234–2250. pmid:34991288
- 20. Guilbert E, Guidoti M. New species and new record of Tingidae (Hemiptera Heteroptera) from Thailand. Zootaxa. 2018; 4482: 274–296.
- 21. Rietschel S. Corythauma ayyari (Drake, 1933)–Erstnachweis der invasiven Tingide in Griechenland (Heteroptera, Tingidae). Heteropteron. 2015; 44: 5–8.
- 22. Sillero N. & Márcia Barbosa A. Common mistakes in ecological niche models, International Journal of Geographical Information Science. 2021; 35: 213–226,
- 23. Zizka A, Silvestro D, Andermann T, Azevedo J, Duarte Ritter C, Edler D, et al. Coordinate cleaner: standardized cleaning of occurrence records from biological collection databases. Methods in Ecology and Evolution. 2019; 10: 744–751.
- 24. Pearson RG, Dawson TP. Predicting the impacts of climate change on the distribution of species: are bioclimate envelope models useful? Global ecology and biogeography. 2003; 12: 361–371.
- 25. Schickele A, Leroy B, Beaugrand G, Goberville E, Hattabe T, Francoura P, et al. Modelling European small pelagic fish distribution: Methodological insights. Ecological Modelling. 2020; 416: 108902.
- 26. Stockwell DRB, Peterson AT. Effets of sample size on accuracy of species distribution models. Ecological Modelling. 2022; 148:1–13.
- 27. Hernandez PA, Graham CH, Master LL, Albert DL. The effect of smple size and species caracteristics on performance of different species distribution modeling methods. Ecography. 2006; 29: 773–785.
- 28. Jiménez-Valverde A. Sample size for the evaluation of presence-absence models. Ecological Indicators. 2020; 114, 106289.
- 29. Voldoire A, Saint-Martin D, Senesi S, Decharme B, Alias A, Chevallier M, et al. Evaluation of CMIP6 DECK experiments with CNRM-CM6-1. Journal of Advances in Modeling Earth Systems. 2019; 11: 2177–2213.
- 30. Thuiller W, Georges D, Engler R, Breiner F. ‘Biomod2’: Ensemble platform for species distribution modelling. 2016; Available from: https://cran.r-project.org/web/packages/biomod2/index.html.
- 31. Araújo MB, New M. Ensemble forecasting of species distributions. Trends in Ecology & Evolution. 2007; 22: 42–47. pmid:17011070
- 32. Allouche O, Tsoar A, Kadmon R. Assessing the accuracy of species distribution models: prevalence, kappa and the true skill statistic (TSS). Journal of Applied Ecology. 2006; 43: 1223–1232.
- 33. Boyce MS, Vernier PR, Nielsen SE, Schmiegelow FK. Evaluating resource selection functions. Ecological Modelling. 2022; 15: 281–300.
- 34. Di Cola V, Broennimann O, Petitpierre B, D’Amen M, Breiner F, Guisan A. Ecospat: an R package to support spatial analyses and modeling of species niches and distributions. Ecography. 2017; 40: 774–787.
- 35. Broennimann O, Petitpierre B, Di Cola V, Breiner F, D’Amen M, Randin C, et al. Package ‘ecospat’: Spatial Ecology Miscellaneous Methods. 2016; Available from https://cran.r-project.org/web/packages/ecospat/index.html.
- 36. Correa Nogueira TDA, Ayala WE, Dayrell JS, de Fraga R, Kaefer IL. Scale-dependent estimates of niche overlap and environmental effects on two sister species of Neotropical snakes. Studies on Neotropical Fauna and Environment. 2019; 54: 121–132.
- 37. Schoener TW. Anolis lizards of Bimini: resource partitioning in a complex fauna. Ecology. 1968; 49: 704–726.
- 38. Beck HE, Zimmermann NE, McVicar TR, Vergopolan N, Berg A, Wood EF. Present and future Köppen-Geiger climate classification maps at 1-km resolution. Scientific data. 2018; 5: 1–12. pmid:30375988
- 39. Kottek M, Grieser J, Beck C, Rudolf B, Rubel F. World map of the Köppen-Geiger climate classification updated. Meteorologische Zeitschrift. 2006; 15: 259–263
- 40. Peel MC, Finlayson BL, McMahon TA. Updated world map of the Köppen-Geiger climate classification. Hydrology and earth system sciences. 2007; 11: 1633–1644.
- 41. Pinto MADS Gonçalves APS, Santos SAP Almeida MRLD, Azevedo JCMD. Invasão biológica de Corythucha ciliata em espaços verdes urbanos de Portugal: modelação do nicho ecológico com o método de máxima entropia. Ciência Florestal. 2014; 24: 597–607.
- 42. Sanchez-Ramos I, Pascual S, Fernandez CE, Marcotegui A, Gonzalez-Nunez M. Effect of temperature on the survival and development of the immature stages of Monosteira unicostata (Hemiptera: tingidae). European Journal of Entomology. 2015; 112.
- 43. Kezik U, Eroglu M. The Damage of Turkey’s new invasive species, Corythucha ciliata (Say, 1832) (Hemiptera: Tingidae) in the Eastern Black Sea Region. In Proceedings of Turkey II. Forest Entomology and Pathology Symposium. 2014; pp. 7–9.
- 44.
Lee JY, Marotzke J, Bala G, Cao L, Corti S, Dunne JP, et al. Future global climate: scenario-based projections and near-term information. 2021; Cambridge University Press online. https://doi.org/10.1017/9781009157896.006
- 45. Hallman TA, Robinson WD. Deciphering ecology from statistical artefacts: Competing influence of sample size, prevalence and habitat specialization on species distribution models and how small evaluation datasets can inflate metrics of performance. Diversity and Distribution. 2020; 26: 315–328.
- 46. Atwater D.Z., Ervine C. & Barney J.N. Climatic niche shifts are common in introduced plants. Nature Ecology and Evolution. 2018; 2: 34–43. pmid:29203919
- 47. Petitpierre B, Kueffer C, Broennimann O, Randin C, Daehler C, Guisan A. Climatic niche shifts are rare among terrestrial plant invaders. Science. 2012; 335(6074): 1344–8. pmid:22422981
- 48. Webber BL, Scott JK. Rapid global change: implications for defining natives and aliens. Global Ecology and Biolgeography. 2011; 21: 205–311.