Next Article in Journal
The Role of Self-Efficacy and Activity Patterns in the Physical Activity Levels of Women with Fibromyalgia
Previous Article in Journal
Biomechanics of Traumatic Head and Neck Injuries on Women: A State-of-the-Art Review and Future Directions
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Biology
Volume 12
Issue 1
10.3390/biology12010084
biology-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles 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

Predicting the Impact of Climate Change on the Habitat Distribution of Parthenium hysterophorus around the World and in South Korea

by
Pradeep Adhikari
1,†,
Yong-Ho Lee
1,2,†,
Anil Poudel
3,
Gaeun Lee
1,
Sun-Hee Hong
3,* and
Yong-Soon Park
4,*
1
Institute of Humanities and Ecology Consensus Resilience Lab, Hankyong National University, Anseong 17579, Republic of Korea
2
OJeong Resilience Institute, Korea University, Seoul 02841, Republic of Korea
3
School of Plant Science and Landscape Architecture, College of Agriculture and Life Sciences, Hankyong National University, Anseong 17579, Republic of Korea
4
Department of Plant Resources, College of Industrial Sciences, Kongju National University, Yesan 32439, Republic of Korea
*
Authors to whom correspondence should be addressed.
†
These authors contributed equally to this work.
Biology 2023, 12(1), 84; https://doi.org/10.3390/biology12010084
Submission received: 14 November 2022 / Revised: 14 December 2022 / Accepted: 27 December 2022 / Published: 4 January 2023
Browse Figures
Versions Notes

Abstract

:

Simple Summary

Parthenium hysterophorus is one of the most noxious invasive weeds in the world. In this study, a species distribution model of P. hysterophorus was established using the maximum entropy (MaxEnt) modeling approach. According to the model, climate change would likely increase the habitat suitability for P. hysterophorus across the world. Estimation of mean habitat suitability revealed that 21 countries including Bulgaria, Brunei, China, Netherlands, New Zealand, and South Korea, currently in the low habitat suitability category, would transition into the moderate to very high suitability category by 2081–2100. In South Korea, climate change would increase the habitat suitability for P. hysterophorus, especially in the southern region of the country, which would be followed by spontaneous expansion towards the northern region, thus seriously threatening agriculture, native biodiversity, ecosystem services, and the national economy. Therefore, regular monitoring is required to perform to prevent further habitat expansion of P. hysterophorus in South Korea.

Abstract

The global climate change, including increases in temperature and precipitation, may exacerbate the invasion by P. hysterophorus. Here, MaxEnt modeling was performed to predict P. hysterophorus distribution worldwide and in South Korea under the current and future climate global climate changes, including increases in temperature and precipitation. Under the current climate, P. hysterophorus was estimated to occupy 91.26%, 83.26%, and 62.75% of the total land area of Australia, South America, and Oceania, respectively. However, under future climate scenarios, the habitat distribution of P. hysterophorus would show the greatest change in Europe (56.65%) and would extend up to 65°N by 2081–2100 in South Korea, P. hysterophorus currently potentially colonizing 2.24% of the land area, particularly in six administrative divisions. In the future, P. hysterophorus would spread rapidly, colonizing all administrative divisions, except Incheon, by 2081–2100. Additionally, the southern and central regions of South Korea showed greater habitat suitability than the northern region. These findings suggest that future climate change will increase P. hysterophorus distribution both globally and locally. Therefore, effective control and management strategies should be employed around the world and in South Korea to restrict the habitat expansion of P. hysterophorus.

1. Introduction

Parthenium hysterophorus, a prolific invasive species, is native to the tropical and subtropical regions of America [1]. P. hysterophorus is an annual herbaceous plant species belonging to the family Asteraceae [2] and is recognized as one of the 100 of the world’s worst invasive species. P. hysterophorus exhibits high fecundity and rapid germination and growth and produces a number of growth-inhibiting phytochemicals that suppress the germination and growth of the surrounding flora [2,3,4]. Moreover, it is highly competitive with pasture plants and food crops for soil nutrients (e.g., nitrogen and soil moisture) [5].
P. hysterophorus can tolerate salt and drought stresses [6] and usually invades disturbed land, but it can adapt to a variety of habitats including croplands, pastures, orchards, and the edges of forests [1], which makes it one of the most noxious invasive weeds in the world, threatening biodiversity and significantly reducing crop and pasture yield [2]. In India, P. hysterophorus has been reported to decrease crop and fodder yield by up to 63% and 90%, respectively [7]. Moreover, P. hysterophorus also has socioeconomic impacts, such as poisoning in livestock and wildlife, and causing pollen allergy, contact dermatitis, asthma, bronchitis, and hay fever in human beings [4,8].
The first taxonomic record of P. hysterophorus was in east Mexico; however, its native range is considered to span the West Indies and the adjoining coastal areas of North and South America [9]. Because of accelerated human mobility, owing to international trade and tourism and technological advancements, P. hysterophorus has moved into other countries of the Americas, and subsequently into Asia, Africa, Australia Oceania, and Europe [2,9]. Currently, P. hysterophorus is found in over 50 countries [4]. In South Korea, P. hysterophorus was first recorded before 1996 in Masan City, which is located along the southeastern coast of the country [10].
P. hysterophorus can tolerate a wide range of environments, including high temperature, extreme soil moisture, and increasing carbon dioxide (CO2) concentration [11], suggesting that P. hysterophorus is a highly prolific plant under climate change. In addition, changes in temperature and precipitation are very critical climatic factors that determine plant distribution and invasiveness [12]. Besides the climatic factors, other environmental components, such as wind, surface runoff, movement of livestock or wildlife, and anthropogenic disturbance, e.g., construction of roads, railways, and parks, influence the distribution of P. hysterophorus [2,3]. P. hysterophorus prefers neutral to alkaline pH soils, but it can grow in all types of soils in fields and wastelands [13]. Therefore, to manage invasive species, conserve native biodiversity, and understand the establishment and spread of invasive species is critical under the current and future climate scenarios [14]. The information gained will enable the early detection of invasive species and will help establish a rapid response system for controlling and eradicating the invasive species [15,16].
Species distribution models (SDMs) are currently the most reliable techniques used by invasion biologists to investigate the impact of climate change on the geographical distribution and range expansion of invasive species [3,14,17,18]. The SDMs correlate the species occurrence with climatic and other environmental variables, based on the principle of niche conservatism, to generate maps displaying the potential distribution of a given species [19]. Among the various SDMs, the maximum entropy (MaxEnt) model is a popular machine-learning technique that can achieve high predictive accuracy based on a small number of species-occurrence records and environmental variables [20,21]. The algorithm has been widely used to predict the current and future habitats of various invasive species [3,16,22].
The points mentioned above suggest that P. hysterophorus is a highly prolific plant under climate change [4] and can be adapted by a variety of environmental factors [23,24]. The average temperature of South Korea has increased by 1.8 °C over the past 100 years, and is predicted to increase by 5.7 °C by 2100 [25]. Therefore, the future climate of South Korea is expected to favor the habitat expansion of P. hysterophorus. In our earlier studies, we assessed the risk of invasive species, including ecosystem-disturbing alien plants (EDAPs), under the changing climate of South Korea [15,17,26]. The results revealed that South Korea is at a high risk of invasion under future climate conditions. However, because of the limited availability of species-occurrence records, we could not perform the modeling of EDAPs, such as P. hysterophorus, in these studies. To the best of our knowledge, no studies have yet been undertaken to reveal the invasion potential of P. hysterophorus in South Korea. Therefore, we collected the global occurrence records of P. hysterophorus and designed this study. The main objectives of this study were as follows: (1) to predict the current and future distribution range of P. hysterophorus, both globally and locally in South Korea, using the MaxEnt algorithm; (2) to evaluate the habitat expansion of P. hysterophorus in different administrative divisions (ADs) of South Korea; and (3) to classify the vulnerability of each AD based on the habitat suitability index. Overall, we identified areas at high risk of invasion by P. hysterophorus. The results of this study enhance our understanding of the current distribution pattern and future expansion potential of P. hysterophorus in South Korea and support the construction of a theoretical framework that could be used to develop management strategies for controlling its further spread.

2. Materials and Methods

2.1. Species-Occurrence Records

A total of 16,353 global species-occurrence records of P. hysterophorus were downloaded from the Global Biodiversity Information Facility [27]. Then, multiple species-occurrence points in the same grid at a spatial resolution of 2.5 min (~4.5 km2) were removed and selected a single unique point per grid by using the spatially rarefy occurrence tool in the ArcGIS SDM toolbox v. 2.4 [28]. This process prevents the overfitting and incorrect inflation of the model outcomes, owing to spatial autocorrelation [29]. Finally, the number of species-occurrence points of P. hysterophorus was reduced to 9234 (Figure 1 and Table S1). Here, both species-occurrence data sets were used in the MaxEnt modeling of P. hysterophorus for investigation of the overestimation of the model.

2.2. Selection of Bioclimatic Variables

Nineteen bioclimatic variables, which were considered to be important for predicting the global distribution pattern of P. hysterophorus, recorded over a period of 30 years (1970–2000) at a spatial resolution of 2.5 min (~4.5 km at the equator), were downloaded from the WorldClim data portal [30]. WorldClim v2.1 was used to project historical (1970–2000) data, which were considered as the current climatic data. Therefore, these data are hereafter referred to as the current climate data. Similarly, future bioclimatic variables, recorded at the same resolution, were also downloaded from the WorldClim data portal using Coupled Model Intercomparison Project Phase 6 (CMIP6) [31]. The global climate model, Max Planck Institute for Meteorology Earth System Model (MPI-ESM1-2-HR) [32], and two shared socioeconomic pathways (SSPs: SSP2-4.5 and SSP5-8.5) were used to represent the climatic data for the future periods of 2021–2040, 2041–2060, 2061–2080, and 2081–2100. The SSPs are scenarios of projected socioeconomic global changes up to 2100. The SSP scenarios evaluate changes in land use and energy consumption, as well as the corresponding uncertainty in the emission of greenhouse gases and air pollutants [33]. Under SSP2-4.5 and SSP5-8.5, the global mean surface temperature was predicted to increase by 1.8–4.1 K and 3.8–8.6 K, respectively, relative to 1750 [34]. The WorldClim data portal has been used widely for downloading bioclimatic variables for predicting the potential distribution of species in response to the changing temperature and precipitation. The bioclimatic variables serve to delineate and predict the future distribution patterns of species, depending on various climate scenarios and their ecologies [35]. To identify the most important variables for the modeling of P. hysterophorus, the data of 19 bioclimatic variables downloaded from the WorldClim data portal (Table S2) were subjected to the Spearman’s correlation test. Six variables, including the annual mean temperature (Bio01), mean diurnal temperature range (Bio2), isothermality (Bio03), annual precipitation (Bio12), precipitation in the wettest month (Bio13), and precipitation in the driest month (Bio 14), were ultimately selected for the MaxEnt modeling (Table 1), based on their weak correlation with each other (r < 0.75; Table S3). These six variables were considered as the most important climatic factors for predicting the occurrence of P. hysterophorus.The Pearson correlation analysis was performed using the PROC CORR function of SAS 9.4 (SAS Institute, Inc., Cary, NC, USA), and six variables were selected, as described previously [36,37] (Table S3). In addition to the bioclimatic variables, other environmental variables, such as land use and landcover change, soil, and human disturbance, could be important variables for determining the distribution of P. hysterophorus, but future data of such variables are not available under a similar resolution.

2.3. Model Development, Evaluation, and Validation

The current and future distribution patterns of P. hysterophorus were investigated worldwide and in different provinces of South Korea using the MaxEnt Package version 1.33 (http://cran.r-project/org/src/contrib/archive/maxent/ (accessed on 1 September 2022). MaxEnt is a popular machine-learning technique for studying habitat suitability that exhibits high predictive performance based on only a few species-occurrence data points [38]. MaxEnt is generally the best technique for studying invasive species, since the lack of data points for such species may not be reliable as their range may be expanding and may have not reached an equilibrium, which may lead to the misinterpretation of habitat suitability [39]. The background data points of the study area were determined using ArcGIS 10.3, as recommended previously [40], and used for the MaxEnt modeling. In this study, 75% of the species-occurrence data points were used for the model calibration, and the remaining 25% were used for the model validation [41]. The other model options were run with the default settings, and the model was replicated 100 times, as described previously [15].
The goodness-of-fit of the model was measured using three evaluation parameters, namely, the area under the curve (AUC) values of the receiver operating characteristic (ROC) curves [42], the true skill statistic (TSS) [43], and the kappa statistic. When testing model results, the AUC is a threshold-independent method for differentiating between presence and absence. The AUC value, which ranges from 0 to 1, assesses the performance of a model [44]. The AUC value is independent of the size of the dataset (prevalence); however, its use is questionable, because it assigns equal weights to both commission and omission errors, which may prevent accurate predictions [45]. Habitat expansion outside of the species-occurrence range may produce high AUC values, leading to overfitting, a phenomenon that misleads model evaluation [46]. To avoid this problem, other evaluation parameters, such as the TSS and kappa statistics, were also used to understand the accuracy of the model. According to the AUC values, the robustness of the model is rated as failed (0.5–0.6), poor (0.6–0.7), fair (0.7–0.8), good (0.8–0.9), and excellent (0.9–1) [47]. The TSS, which assesses both the specificity and sensitivity of the model, ranges from −1 to +1, accounting for both omission and commission errors [43], and is used as an alternative for assessing the model accuracy [43,48]. The kappa statistic measures the accuracy of the predictions in relation to what may have been discovered by chance alone [43]. Like the TSS, the kappa statistic also ranges from −1 (poor agreement) to +1 (perfect prediction) [43].

2.4. Habitat Expansion of P. hysterophorus across the World and in South Korea

The probability distribution map obtained from the MaxEnt modeling was used to produce binary distribution maps of P. hysterophorus on a global scale, under the threshold maximum training sensitivity plus specificity cloglog [49], under the current and future climate change scenarios (SSP2-4.5 and SSP5-8.5) for the periods of 2021–2040, 2041–2060, 2061–2080, and 2081–2100. The binary distribution maps show suitable and unsuitable habitats for P. hysterophorus. The number of habitat-suitable cells for P. hysterophorus were estimated to determine the rate of its habitat expansion on different continents, nations (global scale), and in different ADs of South Korea (local scale) using the Spatial Analyst tool in ArcGIS Desktop 10.8 (Esri, Redlands, CA, USA). The approximate area (1 grid cell = 4.5 km2), changes in the habitat suitability, and mean habitat suitability for P. hysterophorus were estimated across the different continents, nations, and local ADs in South Korea. Based on the mean suitable value, habitat suitability was classified into four categories, namely, low (≤0.25), moderate (0.26–0.50), high (0.51–0.75), and very high (≥0.76). Similarly, the habitat suitability of 17 ADs in South Korea was classified into the low, moderate, high, and very high categories, based on the mean score of habitat suitability under the current and future climate scenarios.

3. Results

3.1. Evaluation of Bioclimatic Variables

To investigate the importance of each of the six variables, we estimated their average contribution to the model over five time periods, including the current period (1973–2000) and four future periods (2021–2040, 2041–2060, 2061–2080, and 2081–2100) (Table S4). Among the six variables selected above, three variables (Bio1, Bio3, and Bio13) showed a relatively high contribution to the model. Among these three variables, Bio1 showed the highest average contribution (40.48%), followed by Bio13 (27.19%) and Bio3 (23.75%) (Table 1). Therefore, these three variables were identified as the most prominent factors driving the distribution of P. hysterophorus; other variables played a relatively smaller contribution in the model’s performance. Similarly, the permutation importance of these bioclimatic variables was estimated, which showed that Bio1, Bio3, and Bio12 are, relatively, the most important variables in the model (Table S5), estimated to be of average importance 35.45%, 21.77%, and 15.18%, respectively. Other variables are relatively less important. The variable importance was further confirmed using the jackknife approach, which estimates the relevance of each variable in the species distribution model and the distinctness of the information provided by each variable. Consistent with the Spearman correlation analysis, the jackknife approach revealed Bio1, Bio3, and Bio13 as the most important variables (Figure S1a–i).

3.2. Evaluation of Model Performance Based on AUC, TSS, and Kappa Scores

The model performance was evaluated based on the AUC, TSS, and kappa scores. The AUC, TSS, and kappa scores were highest in the MaxEnt modeling using rarified species-occurrence points compared to the MaxEnt modeling using all the occurrence points (Table 2). It indicates that the model predictions from the rarified species-occurrence points are relatively more accurate and less likely to overestimate the model. Therefore, the model outputs obtained from the rarified occurrence points were selected in this study. The AUC value of the selected MaxEnt model was 0.776, which implies good model performance. Similarly, the TSS and kappa scores were 0.788 and 0.685, respectively, indicating agreement between the observed and predicted data of the model. These results confirmed that the model’s performance when predicting the spatial distribution of P. hysterophorus, based on the presence-only data, was excellent.

3.3. Impact of Climate Change on P. hysterophorus Distribution Worldwide

MaxEnt modeling was performed to examine the current and future distribution patterns of P. hysterophorus under the current climate (1973–2000) and future climate scenarios (SSP2-4.5 and SSP5-8.5; 2021–2040, 2041–2060, 2061–2080, and 2081–2100) (Figure 2 and Figure 3). Under the current climate, P. hysterophorus shows widespread potential distribution in Australia, South America, Oceania, and Africa, covering approximately 91.26%, 83.26%, 62.75%, and 61.29% of the total land surface of each continent, respectively (Table S6). However, according to the model-based predictions, Europe would be the new invasion hotspot of P. hysterophorus in the future. A comparison of future and current habitat suitability indices revealed that the greatest change in P. hysterophorus habitat would be observed in Europe (SSP2-4.5= 61.5%, SSP5-8.5 = 56.65%), followed by Oceania (SSP2-4.5 = 34.30%, SSP5-8.5 = 26.76%), and Asia (SSP2-4.5 = 30.92%, SSP5-8.5 = 38.12%), by 2081–2100 (Table 3). These results indicate that the habitat of P. hysterophorus would expand in the northern regions of the world, extending up to a latitude of 65°N.
Similarly, the mean habitat suitability for P. hysterophorus in all countries around the world was estimated and classified into low, moderate, high, and very high. Under the current climate, 81 countries showed low habitat suitability (Figure S2 and Table S7). However, of these 81 countries, 3 countries (Monaco, Netherlands, and Tokelau (dependent New Zealand territory)) under SSP2-4.5 and 5 countries (Djibouti, Monaco, Netherlands, South Korea, and Tokelau NZ under SSP5-8.5 would transition into the very high suitability category by 2081–2100 (Table 4). These results indicate that climate change would facilitate the expansion of the P. hysterophorus habitat on the global scale.

3.4. Impact of Climate Change on P. hysterophorus Distribution in South Korea

The species distribution model showed that P. hysterophorus will undergo extreme expansion in South Korea in the future. The current and future distribution patterns of P. hysterophorus in different time periods are presented in Figure 2 and Figure 3.
Based on our model predictions, the current potential distribution of P. hysterophorus is estimated to be 558 km2 in South Korea, covering 2.24% of the total land area of the country (Figure 4). However, in the 2041–2060, 2061–2080, and 2081–2100 periods, the distribution of P. hysterophorus will increase by 405.64%, 2,491.12%, and 2,835.48%, respectively, under SSP2-4.5, and by 566.93%, 2,883.06%, and 3,343.54%, respectively, under SSP5-8.5. These results suggest that South Korea will be the future hotspot of P. hysterophorus.
Next, we evaluated the habitat suitability for P. hysterophorus in the different ADs of South Korea. Table 5 shows the estimated area of P. hysterophorus under the current and future climate scenarios in 17 ADs. Currently, P. hysterophorus exists in six ADs, including South Jeolla, North Jeolla, South Chungcheong Gwangju, North Gyeongsang, and Daegue, with each AD estimated to be 18–220.5 km2 (Table 5). However, the suitable habitats for P. hysterophorus would expand across the whole country by 2081–2100 in all ADs (except Incheon and Seoul, located in the northwestern part of the country) under SSP2-4.5 and only in Incheon under SSP5-8.5. By 2081–2100, the abundance of P. hysterophorus would be the highest in North Gyeongsang (4599 km2).
The suitable habitats for P. hysterophorus were classified into four categories (low, moderate, high, and very high), based on the mean habitat suitability measured on a linear scale (Figure 5 and Figure 6). Under the current climate, the habitat suitability for P. hysterophorus was estimated to be high in Gwangju, and moderate in North Jeolla and Daegue. However, habitat suitability would be high in Jeju and South Chungcheong, and very high in 12 ADs (e.g., South Jeolla, North Chungcheong, and North Gyeongsang) under SSP2-4.5, and very high in 13 ADs under SSP5-8.5 by 2081–2100. These results suggest that climate change will likely promote the northward expansion of P. hysterophorus habitats in South Korea.

4. Discussion

The main findings of this study are as follows: First, among the six bioclimatic variables, the annual mean temperature was the most important variable affecting the global distribution of P. hysterophorus (Table 1). Second, the model predictions from the rarified species-occurrence points are relatively more accurate and less likely to overestimate the MaxEnt model. Third, under the current climate, Australia showed the highest proportion of suitable habitats relative to its total land area; however, the habitat suitability for P. hysterophorus is predicted to be the highest in Europe by 2081–2100. Fourth, under the current climate, 81 countries showed low habitat suitability (Table S7); however, under future climate scenarios, five countries, including Djibouti, Monaco, Netherlands, South Korea, and Tokelau (New Zealand), would transition from the low suitability category to the very high suitability category by 2081–2100 (Table 4). Fifth, only 2.24% of the area of South Korea is suitable for P. hysterophorus under the current climate, but the area covered by P. hysterophorus is predicted to increase by approximately 3343.54% by 2081–2100 under SSP5-8.5, making South Korea the future hotspot of this invasive species. Lastly, the ADs in the northern and northwestern regions of South Korea are relatively safe from the invasion of P. hysterophorus compared with those in the southern region of the country (Figure 5 and Figure 6).
The SDM technique is used globally to assess the invasion risk of alien species [50]. Several factors control the performance, and consequently the output, of the model including the size and resolution of the study site, threshold used for modeling [51], type of species, number of species-occurrence records, the type and number of variables selected [52,53], modeling approach [54], global circulation models used [55], methods of model evaluation and validation [46], and thresholds used in binary distribution maps [56]. Among the various algorithms, the MaxEnt model is well known for the modeling of invasive species, because it exhibits high predictive performance using presence-only data [20]; the absence data of invasive species are usually not available, because the ecological ranges of the species may be expanding and may not have reached equilibrium, which may affect the modeling accuracy [39]. Moreover, the MaxEnt model can run easily with many default options [20]. Therefore, we decided to use the MaxEnt model in this study. The predictive performance of this model was tested using the AUC, TSS, and kappa scores, because the use of only the AUC score for the selection of a model is not recommended [43]. In the present study, the AUC, TSS, and kappa scores were calculated as 0.776, 0.788, and 0.685, respectively, which indicated that the model used to study the potential distribution of P. hysterophorus was robust and highly accurate. Moreover, the MaxEnt model may have overfitting issues due to the sampling bias in geographic and environmental space [29], which could be reduced based on the selection of species-occurrence points. In this study, as a preliminary test, we addressed the issue of overfitting using the rarified species-occurrence points in the model.
The MaxEnt modeling of P. hysterophorus was performed using six important bioclimatic variables. Among these variables, those related to temperature (Bio1 and Bio3) and precipitation (Bio13) significantly affected the global distribution of P. hysterophorus. This finding is consistent with some recent studies, which used temperature- and precipitation-related variables as the primary factors determining the distribution of invasive species [3,14,18]. Temperature and precipitation play pivotal roles in plant physiological activities such as germination, growth, and development [57]. P. hysterophorus can germinate in a wide temperature range (4–36 °C) [23] and can resist drought stress [5]. Moreover, increasing atmospheric CO2 concentration increases herbicide, e.g., glyphosate, tolerance capacity and may hinder chemical control effects against P. hysterophorus [24]. Therefore, future climate change favors the habitat expansion of P. hysterophorus.
The habitat suitability for P. hysterophorus not only depends on the bioclimatic variables used in the model but also on many non-climatic factors, including habitat characteristics, land topography, and its superior morphological and physiological characteristics, such as a short life cycle, high fecundity (as evident from the production of 15,000–156,768 viable seeds per specimen [58]), strong dispersal ability, persistent soil seed bank, high germination and growth rates, and high environmental stress tolerance [2]. Therefore, P. hysterophorus has a strong competitive ability [2,3,4] and can occupy empty niches and substantially expand in new geographical areas [3,14,59,60]. Similarly, many studies showed that disturbed lands, such as roadsides, urban areas, parks, abandoned farms, and orchards, were ideal habitats for P. hysterophorus [3].
Our study revealed that the global distribution of P. hysterophorus under the current climate is concentrated in tropical and subtropical regions located in the equatorial belt (35°N to 35°S of the equator), which includes Central America, the Amazon basin of South America, the Congo basin of Africa, and the Indomalayan region of Asia and Australasia. These results are consistent with some recent studies [3,61]. However, by 2081–2100, the global distribution of P. hysterophorus could expand up to 60°N, and many countries of Europe such as Ireland, the Netherlands, and United Kingdom could become its new hotspots (Table S7). This increase in habitat suitability for P. hysterophorus in many countries in the Northern Hemisphere could be attributed to climate change [3,4].
Similarly, the mean habitat suitability for P. hysterophorus in each country was estimated and classified into four categories (low, moderate, high, and very high) (Figure S2 and Table S7). Among the countries in the low habitat suitability group under the current climate, eight countries, including China, Belgium, and Syria, would transition into the moderate category by 2081–2100; six countries, including Brunei, Bulgaria, and Slovania, into the high habitat suitability category; and five countries, including South Korea, Monaco, and the Netherlands, into the very high habitat suitability group under the SSP5-8.5. However, such changes could not be detected in the global prediction maps because of the high-resolution scale of 2.5 min (~4.5 km2) (Figure 2 and Figure 3). The mean habitat suitability was found to be higher under SSP5-8.5 than SSP2-4.5, particularly in the Northern Hemisphere above 30°N latitude. Nonetheless, there were no significant differences from the equator to 15°N latitude (e.g., Colombia, Venezuela, Senegal, Ethiopia, India, and Thailand) or in the Southern Hemisphere (for example, Argentina, Brazil, Zambia, Madagascar, and Australia). This indicates that P. hysterophorus can tolerate a wide range of temperatures, and even a low magnitude of warming will allow major expansion of this species. These results are the outcomes of global climate change, increasing the growth and reproduction capacity under elevated CO2 levels, higher temperature, and drought [4,11,62], leading to the range expansion of P. hysterophorus. These results are useful in designing country-specific profiles, depending on the invasion risk category. Furthermore, climate change will create suitable ecological niches for P. hysterophorus at higher elevations. Although this study did not show the elevational range shift, two separate studies in Bhutan and Nepal showed that the habitat of P. hysterophorus extends up to 2931 m above sea level (masl) [63,64]. These results suggest that climate change would facilitate the habitat expansion of P. hysterophorus to higher altitudes and latitudes, owing to the increase in temperature.
In South Korea, our model predicted that small areas of the country are climatically suitable for P. hysterophorus under the current conditions. Total suitable habitat area was estimated under the current conditions. Our model predicted that a total of 558 km2 in six ADs would be invaded by P. hysterophorus under the current climate. Three of these six ADs including North Jeolla, South Jeolla, and North Gyeongsang, which are located in the western, southern, and eastern regions of South Korea, respectively, have been invaded to a greater extent than the other three ADs. All of the invaded ADs of South Korea are located below 36°N, and these regions are characterized by a warm temperate climate with high humidity [25]. These conditions may favor the establishment of not only P. hysterophorus but also many other invasive weeds found in the tropical and subtropical regions of the Americas, Southern Europe, and East and South Asia [1,2,10]. The invasive weeds native to tropical and subtropical climates exhibit relatively higher critical thermal maxima than indigenous species, suggesting that such invasive species can flourish at higher temperatures and can dominate the indigenous species under the changing climate [60].
Under the future climate, the invasion by P. hysterophorus would substantially increase and expand northward. By 2081–2100, the P. hysterophorus habitat would expand in all the ADs of South Korea, except Seoul and Incheon, which are located above 37°N in the northwestern region of the country. These results complement the spatially explicit evidence, thus supporting the earlier hypothesis, according to which increasing temperatures are expected to expand invasion threats in the northward direction [65], consistent with previous reports [15,63,64,66]. Moreover, our study showed that the ADs present in the southern and central regions of South Korea, except Gwangju, with a mean habitat suitability ranging from low to high in 2041–2060, would advance to the very high suitability category by 2081–2100 (Figure 6). With the rising global temperature, the habitat suitability for P. hysterophorus is predicted to expand towards the central and northern regions, similar to that of other invasive species in South Korea [15,16,22], because of the elimination of current climatic barriers, thus impelling the establishment of plant hardiness zones northward [60,67].
Besides climate change, anthropogenic factors such as the construction of roads and railway networks could facilitate the invasion of adjacent crop fields and pastures [68]. In South Korea, many highways and railroads connect the north to the south, which may assist in the habitat expansion of P. hysterophorus. Similarly, natural phenomena, such as typhoon, wind, water, and wild animals, as well as human activities such as the import of contaminated grains, vegetables, and pasture seeds may promote invasion [4]. South Korea also experiences such natural phenomena and human activities, and therefore is not immune to invasion by P. hysterophorus in the future.
P. hysterophorus has detrimental impacts on agricultural and natural ecosystems by decreasing crop yield, degrading pastures, and reducing the forage of livestock and wild herbivores (e.g., roe deer [69]), and threatening forest ecosystems [70]. Agricultural land in the western, southern, and eastern regions of South Korea are under high risk of invasion of P. hysterophorus, which may cause high economic losses and negatively impact food security, native biodiversity, and ecosystem services in the country. Therefore, strict quarantine measures are required to limit the habitat expansion of P. hysterophorus, both globally and locally. Additionally, policymakers, land resource managers, and local stakeholders need to develop invasion control strategies to prevent further invasions.

5. Conclusions

P. hysterophorus is a major threat to agriculture, biodiversity, natural ecosystems, and, consequently, to the national economy. To minimize the negative impact of P. hysterophorus, we selected the MaxEnt model to predict its potential distribution around the world and in South Korea. Our findings suggest that P. hysterophorus is distributed between 35°N and 35°S of the equator under the current climate, and its distribution is relatively high in Australia, South America, Oceania, and Africa. However, under future climate scenarios (SSP2-4.5 and SSP5-8.5), the habitat of P. hysterophorus would expand extensively (up to 65° N of the equator), while retaining its current distribution range, and Europe would be the new invasion hotspot. In South Korea, a small portion of the country (2.24%) is predicted to be invaded by P. hysterophorus, particularly in six ADs (mainly North Jeolla, South Jeolla, and North Gyeongsang) under the current climate. However, in the future, the habitat of P. hysterophorus would expand in all ADs, except Incheon, which is located in the northwestern part of the country. Similarly, an estimation of the mean habitat suitability revealed that the central and southern regions of the country would be very highly suitable for P. hysterophorus invasion. Therefore, careful planning and long-term management strategies are needed at the global, national, and local levels to reduce the habitat expansion of this noxious weed. The MaxEnt modeling approach can assist the government in prioritizing the geographical locations for early detection and eradication and can suggest the best preventive measures.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/biology12010084/s1, Table S1: Global occurrence points of P. hysterophorus; Table S2: List of bioclimatic variables; Table S3: Estimation of Pearson correlation coefficient for selecting bioclimatic variables; Table S4: Contribution of different bioclimatic variables in model performance corresponding to different time periods; Table S5: Permutation importance of different bioclimatic variables in model; Table S6: Estimation of suitable habitat area (km2) of P. hysterophorus in different continents under the current (1973–2000) and future climate scenarios SSP2-4.5 and SSP5-8.5; Table S7: Estimation of approximate area of suitable habitat (km2) and mean habitat suitability for P. hysterophorus in different nations of the world under the current climate and future climate scenarios, SSP2-4.5 and SSP5-8.5; Figure S1: Jackknife test-based analysis of the model used to predict the habitat expansion of P. hysterophorus under the current climate (1973–2000) and future climate scenarios (SSP2-4.5 and SSP5-8.5; 2021–2040, 2041–2060, 2061-2080, and 2081–2100). Figure S2: Estimation of the mean habitat suitability for P. hysterophorus in different countries of the world under the current climate (1973–2000) and future climate scenarios (SSP2-4.5 and SSP5; 2021–2040, 2041–2060, 2061–2080, and 2081–2100).

Author Contributions

Conceptualization, P.A., Y.-H.L. and S.-H.H.; methodology, P.A.; software, P.A.; validation, P.A. and Y.-H.L.; formal analysis, P.A. and A.P. investigation, P.A., Y.-H.L. and Y.-S.P.; resources, S.-H.H.; data curation, P.A., A.P. and G.L.; writing—original draft preparation, P.A.; writing—review and editing, P.A. and Y.-S.P.; supervision, Y.-S.P. and S.-H.H.; project administration, S.-H.H.; funding acquisition, S.-H.H. All the authors have read and agreed to the published version of the manuscript. All authors have read and agreed to the published version of the manuscript.

Funding

This study was supported by the Ministry of Environment, Republic of Korea (grant number 2021002270004).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Not applicable.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Adkins, S.W.; Shabbir, A.; Dhileepan, K. Parthenium Weed: Biology, Ecology and Management; CABI: Wallingford, UK, 2018; Volume 7. [Google Scholar]
  2. Adkins, S.; Shabbir, A. Biology, ecology and management of the invasive parthenium weed (Parthenium hysterophorus L.). Pest Manag. Sci. 2014, 70, 1023–1029. [Google Scholar] [CrossRef]
  3. Al Ruheili, A.M.; Al Sariri, T.; Al Subhi, A.M. Predicting the potential habitat distribution of parthenium weed (Parthenium hysterophorus) globally and in Oman under projected climate change. J. Saudi Soc. Agric. Sci. 2022, 21, 469–478. [Google Scholar] [CrossRef]
  4. Mao, R.; Shabbir, A.; Adkins, S. Parthenium hysterophorus: A tale of global invasion over two centuries, spread and prevention measures. J. Environ. Manag. 2021, 279, 111751. [Google Scholar] [CrossRef] [PubMed]
  5. Islam, A. Nitrogen Exploitation Capacity of Parthenium Weed and Its Inhibitory Effects on Growth and Development of Rice. M.S. Thesis, Bangladesh Agricultural University, Ymensingh, Bangladesh, 2010. [Google Scholar]
  6. Ahmad, J.; Bashir, H.; Bagheri, R.; Baig, A.; Al-Huqail, A.; Ibrahim, M.M.; Qureshi, M.I. Drought and salinity induced changes in ecophysiology and proteomic profile of Parthenium hysterophorus. PLoS ONE 2017, 12, e0185118. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  7. Kumar, S. Current spread, impact and management of Parthenium weed in India. Int. Parthenium News. Trop. Sub-Trop. Weed Res. Unit Univ. Qld. Aust. 2012, 5, 1–13. [Google Scholar]
  8. Allan, S.; Shi BoYang, S.B.; Adkins, S.W. Impact of parthenium weed on human and animal health. In Parthenium Weed: Biology, Ecology and Management; CABI: Wallingford, UK, 2019; pp. 105–130. [Google Scholar]
  9. Wey, P. Parthenium hysterophorus (Parthenium Weed); CABI Compendium: Wallingford, UK, 2019. [Google Scholar] [CrossRef]
  10. Jung, S.Y.; Lee, J.W.; Shin, H.T.; Kim, S.J.; An, J.B.; Heo, T.I.; Chung, J.M.; Cho, Y.C. Invasive Alien Plants in South Korea; Korea National Arboretum: Pocheon, Republic of Korea, 2017. [Google Scholar]
  11. Navie, S.; McFadyen, R.; Panetta, F.; Adkins, S. The effect of CO2 enrichment on the growth of a C3 weed (Parthenium hysterophorus L.) and its competitive interaction with a C4 grass (Cenchrus ciliaris L.). Plant Prot. Q. 2005, 20, 61–66. [Google Scholar]
  12. Bradley, B.A.; Blumenthal, D.M.; Wilcove, D.S.; Ziska, L.H. Predicting plant invasions in an era of global change. Trends Ecol. Evol. 2010, 25, 310–318. [Google Scholar] [CrossRef]
  13. Guyana, P.; Paraguay, S. Parthenium hysterophorus L. Asteraceae–Parthenium weed. Bull. OEPP/EPPO Bull. 2014, 44, 474–478. [Google Scholar]
  14. Ahmad, R.; Khuroo, A.A.; Hamid, M.; Charles, B.; Rashid, I. Predicting invasion potential and niche dynamics of Parthenium hysterophorus (Congress grass) in India under projected climate change. Biodivers. Conserv. 2019, 28, 2319–2344. [Google Scholar] [CrossRef]
  15. Hong, S.H.; Lee, Y.H.; Lee, G.; Lee, D.-H.; Adhikari, P. Predicting impacts of climate change on northward range expansion of invasive weeds in South Korea. Plants 2021, 10, 1604. [Google Scholar] [CrossRef]
  16. Adhikari, P.; Jeon, J.-Y.; Kim, H.W.; Shin, M.-S.; Adhikari, P.; Seo, C. Potential impact of climate change on plant invasion in the Republic of Korea. J. Ecol. Environ. 2019, 43, 36. [Google Scholar] [CrossRef]
  17. Adhikari, P.; Lee, Y.H.; Park, Y.-S.; Hong, S.H. Assessment of the spatial invasion risk of intentionally introduced alien plant species (IIAPS) under environmental change in South Korea. Biology 2021, 10, 1169. [Google Scholar] [CrossRef] [PubMed]
  18. Masum, S.M.; Halim, A.; Mandal, M.S.H.; Asaduzzaman, M.; Adkins, S. Predicting Current and Future Potential Distributions of Parthenium hysterophorus in Bangladesh Using Maximum Entropy Ecological Niche Modelling. Agronomy 2022, 12, 1592. [Google Scholar] [CrossRef]
  19. Waldock, C.; Stuart-Smith, R.D.; Albouy, C.; Cheung, W.W.; Edgar, G.J.; Mouillot, D.; Tjiputra, J.; Pellissier, L. A quantitative review of abundance-based species distribution models. Ecography 2022, 2022. [Google Scholar] [CrossRef]
  20. Phillips, S.J.; Anderson, R.P.; Schapire, R.E. Maximum entropy modeling of species geographic distributions. Ecol. Model. 2006, 190, 231–259. [Google Scholar] [CrossRef] [Green Version]
  21. Phillips, S.J.; Dudík, M. Modeling of species distributions with Maxent: New extensions and a comprehensive evaluation. Ecography 2008, 31, 161–175. [Google Scholar] [CrossRef]
  22. Adhikari, P.; Kim, B.-J.; Hong, S.-H.; Lee, D.-H. Climate change induced habitat expansion of nutria (Myocastor coypus) in South Korea. Sci. Rep. 2022, 12, 3300. [Google Scholar] [CrossRef]
  23. Williams, J.; Groves, R. The influence of temperature and photoperiod on growth and development of Parthenium hysterophorus L. Weed Res. 1980, 20, 47–52. [Google Scholar] [CrossRef]
  24. Cowie, B.W.; Venter, N.; Witkowski, E.T.; Byrne, M.J. Implications of elevated carbon dioxide on the susceptibility of the globally invasive weed, Parthenium hysterophorus, to glyphosate herbicide. Pest Manag. Sci. 2020, 76, 2324–2332. [Google Scholar] [CrossRef]
  25. KMA. Korean Climate Change Assessment Report; Korea Meteorological Administration: Sejong, Republic of Korea, 2020; p. 48. [Google Scholar]
  26. Adhikari, P.; Lee, Y.H.; Adhikari, P.; Hong, S.H.; Park, Y.-S. Climate change-induced invasion risk of ecosystem disturbing alien plant species: An evaluation using species distribution modeling. Front. Ecol. Evol. 2022, 10, 880987. [Google Scholar] [CrossRef]
  27. GBIF. GBIF.org Occurrence Download. Available online: https://www.gbif.org/occurrence/download/0201824-220831081235567 (accessed on 8 December 2022).
  28. Brown, J.L.; Bennett, J.R.; French, C.M. SDMtoolbox 2.0: The next generation Python-based GIS toolkit for landscape genetic, biogeographic and species distribution model analyses. PeerJ 2017, 5, e4095. [Google Scholar] [CrossRef] [Green Version]
  29. Boria, R.A.; Olson, L.E.; Goodman, S.M.; Anderson, R.P. Spatial filtering to reduce sampling bias can improve the performance of ecological niche models. Ecol. Model. 2014, 275, 73–77. [Google Scholar]
  30. Fick, S.E.; Hijmans, R.J. WorldClim 2: New 1-km spatial resolution climate surfaces for global land areas. Int. J. Climatol. 2017, 37, 4302–4315. [Google Scholar] [CrossRef]
  31. Eyring, V.; Bony, S.; Meehl, G.A.; Senior, C.A.; Stevens, B.; Stouffer, R.J.; Taylor, K.E. Overview of the coupled model intercomparison project phase 6 (CMIP6) experimental design and organization. Geosci. Model Dev. 2016, 9, 1937–1958. [Google Scholar] [CrossRef] [Green Version]
  32. Gutjahr, O.; Putrasahan, D.; Lohmann, K.; Jungclaus, J.H.; von Storch, J.-S.; Brüggemann, N.; Haak, H.; Stössel, A. Max planck institute earth system model (MPI-ESM1. 2) for the high-resolution model intercomparison project (HighResMIP). Geosci. Model Dev. 2019, 12, 3241–3281. [Google Scholar] [CrossRef] [Green Version]
  33. Riahi, K.; Van Vuuren, D.P.; Kriegler, E.; Edmonds, J.; O’neill, B.C.; Fujimori, S.; Bauer, N.; Calvin, K.; Dellink, R.; Fricko, O. The shared socioeconomic pathways and their energy, land use, and greenhouse gas emissions implications: An overview. Glob. Environ. Chang. 2017, 42, 153–168. [Google Scholar] [CrossRef] [Green Version]
  34. Meinshausen, M.; Nicholls, Z.R.; Lewis, J.; Gidden, M.J.; Vogel, E.; Freund, M.; Beyerle, U.; Gessner, C.; Nauels, A.; Bauer, N. The shared socio-economic pathway (SSP) greenhouse gas concentrations and their extensions to 2500. Geosci. Model Dev. 2020, 13, 3571–3605. [Google Scholar] [CrossRef]
  35. Zhang, Y.; Tang, J.; Ren, G.; Zhao, K.; Wang, X. Global potential distribution prediction of Xanthium italicum based on Maxent model. Sci. Rep. 2021, 11, 16545. [Google Scholar] [CrossRef]
  36. Shin, M.-S.; Seo, C.; Lee, M.; Kim, J.-Y.; Jeon, J.-Y.; Adhikari, P.; Hong, S.-B. Prediction of Potential Species Richness of Plants Adaptable to Climate Change in the Korean Peninsula. J. Environ. Impact Assess. 2018, 27, 562–581. [Google Scholar] [CrossRef]
  37. Adhikari, P.; Kim, H.W.; Shin, M.S.; Hong, S.H.; Cho, Y. Potential distribution of the silver stripped skipper (Leptalina unicolor) and maiden silvergrass (Miscanthus sinensis) under climate change in South Korea. Entomol. Res. 2022, 52, 483–492. [Google Scholar] [CrossRef]
  38. Elith, J.; H. Graham, C.; P. Anderson, R.; Dudík, M.; Ferrier, S.; Guisan, A.; J. Hijmans, R.; Huettmann, F.; R. Leathwick, J.; Lehmann, A.; et al. Novel methods improve prediction of species’ distributions from occurrence data. Ecography 2006, 29, 129–151. [Google Scholar] [CrossRef] [Green Version]
  39. Jiménez-Valverde, A.; Peterson, A.T.; Soberón, J.; Overton, J.M.; Aragón, P.; Lobo, J.M. Use of niche models in invasive species risk assessments. Biol. Invasions 2011, 13, 2785–2797. [Google Scholar] [CrossRef]
  40. Barbet-Massin, M.; Jiguet, F.; Albert, C.H.; Thuiller, W. Selecting pseudo-absences for species distribution models: How, where and how many? Methods Ecol. Evol. 2012, 3, 327–338. [Google Scholar] [CrossRef]
  41. Araujo, M.B.; Guisan, A. Five (or so) challenges for species distribution modelling. J. Biogeogr. 2006, 33, 1677–1688. [Google Scholar]
  42. Pearson, R.G. Species’ distribution modeling for conservation educators and practitioners. Lessons Conserv. 2010, 3, 54–89. [Google Scholar]
  43. Allouche, O.; Tsoar, A.; Kadmon, R. Assessing the accuracy of species distribution models: Prevalence, kappa and the true skill statistic (TSS). J. Appl. Ecol. 2006, 43, 1223–1232. [Google Scholar] [CrossRef]
  44. Thuiller, W.; Lavorel, S.; Araújo, M.B. Niche properties and geographical extent as predictors of species sensitivity to climate change. Glob. Ecol. Biogeogr. 2005, 14, 347–357. [Google Scholar] [CrossRef]
  45. Fielding, A.H.; Haworth, P.F. Testing the generality of bird-habitat models. Conserv. Biol. 1995, 9, 1466–1481. [Google Scholar] [CrossRef]
  46. Lobo, J.M.; Jiménez-Valverde, A.; Real, R. AUC: A misleading measure of the performance of predictive distribution models. Glob. Ecol. Biogeogr. 2008, 17, 145–151. [Google Scholar] [CrossRef]
  47. Swets, J.A. Measuring the accuracy of diagnostic systems. Science 1988, 240, 1285–1293. [Google Scholar] [CrossRef] [Green Version]
  48. Shabani, F.; Kumar, L.; Ahmadi, M. Assessing accuracy methods of species distribution models: AUC, specificity, sensitivity and the true skill statistic. Glob. J. Hum.-Soc. Sci. Res. 2018, 18, 6–18. [Google Scholar]
  49. Liu, C.; Newell, G.; White, M. On the selection of thresholds for predicting species occurrence with presence-only data. Ecol. Evol. 2016, 6, 337–348. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  50. El-Barougy, R.F.; Dakhil, M.A.; Halmy, M.W.; Gray, S.M.; Abdelaal, M.; Khedr, A.-H.A.; Bersier, L.-F. Invasion risk assessment using trait-environment and species distribution modelling techniques in an arid protected area: Towards conservation prioritization. Ecol. Indic. 2021, 129, 107951. [Google Scholar] [CrossRef]
  51. Vale, C.G.; Tarroso, P.; Brito, J.C. Predicting species distribution at range margins: Testing the effects of study area extent, resolution and threshold selection in the Sahara–Sahel transition zone. Divers. Distrib. 2014, 20, 20–33. [Google Scholar] [CrossRef]
  52. Guisan, A.; Graham, C.H.; Elith, J.; Huettmann, F.; Group, N.S.D.M. Sensitivity of predictive species distribution models to change in grain size. Divers. Distrib. 2007, 13, 332–340. [Google Scholar] [CrossRef]
  53. Stankowski, P.A.; Parker, W.H. Species distribution modelling: Does one size fit all? A phytogeographic analysis of Salix in Ontario. Ecol. Model. 2010, 221, 1655–1664. [Google Scholar] [CrossRef]
  54. Kaky, E.; Nolan, V.; Alatawi, A.; Gilbert, F. A comparison between Ensemble and MaxEnt species distribution modelling approaches for conservation: A case study with Egyptian medicinal plants. Ecol. Inform. 2020, 60, 101150. [Google Scholar] [CrossRef]
  55. Steen, V.; Sofaer, H.R.; Skagen, S.K.; Ray, A.J.; Noon, B.R. Projecting species’ vulnerability to climate change: Which uncertainty sources matter most and extrapolate best? Ecol. Evol. 2017, 7, 8841–8851. [Google Scholar] [CrossRef]
  56. Liu, C.; White, M.; Newell, G. Selecting thresholds for the prediction of species occurrence with presence-only data. J. Biogeogr. 2013, 40, 778–789. [Google Scholar] [CrossRef]
  57. Khaeim, H.; Kende, Z.; Jolánkai, M.; Kovács, G.P.; Gyuricza, C.; Tarnawa, Á. Impact of temperature and water on seed germination and seedling growth of maize (Zea mays L.). Agronomy 2022, 12, 397. [Google Scholar] [CrossRef]
  58. Dhileepan, K. Reproductive variation in naturally occurring populations of the weed Parthenium hysterophorus (Asteraceae) in Australia. Weed Sci. 2012, 60, 571–576. [Google Scholar] [CrossRef]
  59. Walther, G.-R.; Post, E.; Convey, P.; Menzel, A.; Parmesan, C.; Beebee, T.J.; Fromentin, J.-M.; Hoegh-Guldberg, O.; Bairlein, F. Ecological responses to recent climate change. Nature 2002, 416, 389–395. [Google Scholar] [CrossRef] [PubMed]
  60. Hellmann, J.J.; Byers, J.E.; Bierwagen, B.G.; Dukes, J.S. Five potential consequences of climate change for invasive species. Conserv. Biol. 2008, 22, 534–543. [Google Scholar] [CrossRef] [PubMed]
  61. Kriticos, D.J.; Brunel, S.; Ota, N.; Fried, G.; Oude Lansink, A.G.; Panetta, F.D.; Prasad, T.R.; Shabbir, A.; Yaacoby, T. Downscaling pest risk analyses: Identifying current and future potentially suitable habitats for Parthenium hysterophorus with particular reference to Europe and North Africa. PLoS ONE 2015, 10, e0132807. [Google Scholar] [CrossRef] [PubMed]
  62. Nguyen, T.; Bajwa, A.A.; Navie, S.; O’donnell, C.; Adkins, S. Parthenium weed (Parthenium hysterophorus L.) and climate change: The effect of CO2 concentration, temperature, and water deficit on growth and reproduction of two biotypes. Environ. Sci. Pollut. Res. 2017, 24, 10727–10739. [Google Scholar] [CrossRef]
  63. Dorji, S.; Lakey, L.; Wangchen, T.; Adkins, S. Predicting the distribution of parthenium weed (Parthenium hysterophorus) under current and future climatic conditions in Bhutan. J. Environ. Occuo. Health. 2021, 12, 169–181. [Google Scholar]
  64. Maharjan, S.; Shrestha, B.B.; Joshi, M.D.; Devkota, A.; Muniappan, R.; Adiga, A.; Jha, P.K. Predicting suitable habitat of an invasive weed Parthenium hysterophorus under future climate scenarios in Chitwan Annapurna Landscape, Nepal. J. Mt. Sci. 2019, 16, 2243–2256. [Google Scholar] [CrossRef]
  65. Catriona, E.; Rogers, J.P.M. Climate change and ecosystems of the Mid-Atlantic Region. Clim. Res. 2000, 14, 235–244. [Google Scholar]
  66. Bradley, B.A.; Wilcove, D.S.; Oppenheimer, M. Climate change increases risk of plant invasion in the Eastern United States. Biol. Invasions 2010, 12, 1855–1872. [Google Scholar] [CrossRef]
  67. Bradley, B.A.; Blumenthal, D.M.; Early, R.; Grosholz, E.D.; Lawler, J.J.; Miller, L.P.; Sorte, C.J.; D’Antonio, C.M.; Diez, J.M.; Dukes, J.S. Global change, global trade, and the next wave of plant invasions. Front. Ecol. Environ. 2012, 10, 20–28. [Google Scholar] [CrossRef] [Green Version]
  68. McDougall, K.L.; Lembrechts, J.; Rew, L.J.; Haider, S.; Cavieres, L.A.; Kueffer, C.; Milbau, A.; Naylor, B.J.; Nuñez, M.A.; Pauchard, A. Running off the road: Roadside non-native plants invading mountain vegetation. Biol. Invasions 2018, 20, 3461–3473. [Google Scholar] [CrossRef] [Green Version]
  69. Adhikari, P.; Park, S.-M.; Kim, T.-W.; Lee, J.-W.; Kim, G.-R.; Han, S.-H.; Oh, H.-S. Seasonal and altitudinal variation in roe deer (Capreolus pygargus tianschanicus) diet on Jeju Island, South Korea. J. Asia-Pac. Biodivers. 2016, 9, 422–428. [Google Scholar] [CrossRef] [Green Version]
  70. Wardle, D.A.; Peltzer, D.A. Impacts of invasive biota in forest ecosystems in an aboveground–belowground context. Biol. Invasions 2017, 19, 3301–3316. [Google Scholar] [CrossRef]
Biology 12 00084 g001 550
Figure 1. Global occurrence of P. hysterophorus (n = 9234 points). The red dots shown in the figure indicate global positioning points of P. hysterophorus.
Figure 1. Global occurrence of P. hysterophorus (n = 9234 points). The red dots shown in the figure indicate global positioning points of P. hysterophorus.
Biology 12 00084 g001
Biology 12 00084 g002 550
Figure 2. Potential distribution of P. hysterophorus around the world and in South Korea under the current climate (1973–2000) and future climate scenario (SSP2-4.5; 2021–2040, 2041–2060, 2061–2080, and 2081–2100). Green and red colors indicate the unsuitable and suitable habitats for P. hysterophorus, respectively.
Figure 2. Potential distribution of P. hysterophorus around the world and in South Korea under the current climate (1973–2000) and future climate scenario (SSP2-4.5; 2021–2040, 2041–2060, 2061–2080, and 2081–2100). Green and red colors indicate the unsuitable and suitable habitats for P. hysterophorus, respectively.
Biology 12 00084 g002
Biology 12 00084 g003 550
Figure 3. Potential distribution of P. hysterophorus around the world and in South Korea under the current climate (1973–2000) and future climate scenarios (SSP5-8.5; 2021–2040, 2041–2060, 2061–2080, and 2081–2100). Green and red colors indicate the unsuitable and suitable habitats for P. hysterophorus, respectively.
Figure 3. Potential distribution of P. hysterophorus around the world and in South Korea under the current climate (1973–2000) and future climate scenarios (SSP5-8.5; 2021–2040, 2041–2060, 2061–2080, and 2081–2100). Green and red colors indicate the unsuitable and suitable habitats for P. hysterophorus, respectively.
Biology 12 00084 g003
Biology 12 00084 g004 550
Figure 4. Potential habitat estimation of P. hysterophorus in South Korea under the current climate (1973–2000) and future climate scenarios (SSP2-4.5 and SSP5-8.5; 2021–2040, 2041–2060, 2061–2080, and 2081–2100). Green and red bars indicate SSP2-4.5 and SSP5-8.5, respectively.
Figure 4. Potential habitat estimation of P. hysterophorus in South Korea under the current climate (1973–2000) and future climate scenarios (SSP2-4.5 and SSP5-8.5; 2021–2040, 2041–2060, 2061–2080, and 2081–2100). Green and red bars indicate SSP2-4.5 and SSP5-8.5, respectively.
Biology 12 00084 g004
Biology 12 00084 g005 550
Figure 5. Mean habitat suitability for P. hysterophorus in different ADs of South Korea under the current climate (1973–2000) and future climate scenarios (SSP2-4.5; 2021–2040, 2041–2060, 2061–2080, and 2081–2100). The mean habitat suitability for P. hysterophorus was classified into four categories (low, moderate, high, and very high), each indicated with a different color.
Figure 5. Mean habitat suitability for P. hysterophorus in different ADs of South Korea under the current climate (1973–2000) and future climate scenarios (SSP2-4.5; 2021–2040, 2041–2060, 2061–2080, and 2081–2100). The mean habitat suitability for P. hysterophorus was classified into four categories (low, moderate, high, and very high), each indicated with a different color.
Biology 12 00084 g005
Biology 12 00084 g006 550
Figure 6. Mean habitat suitability for P. hysterophorus in different ADs of South Korea under the current climate (1973–2000) and future climate scenarios (SSP5-8.5; 2021–2040, 2041–2060, 2061-2080, and 2081–2100). Mean habitat suitability was classified into four categories (low, moderate, high, and very high), each of which is indicated with a different color.
Figure 6. Mean habitat suitability for P. hysterophorus in different ADs of South Korea under the current climate (1973–2000) and future climate scenarios (SSP5-8.5; 2021–2040, 2041–2060, 2061-2080, and 2081–2100). Mean habitat suitability was classified into four categories (low, moderate, high, and very high), each of which is indicated with a different color.
Biology 12 00084 g006
Table
Table 1. Bioclimatic variables selected for modeling P. hysterophorus distribution.
Table 1. Bioclimatic variables selected for modeling P. hysterophorus distribution.
CodeDescriptionUnitModel Contribution (%) 1
Bio1Annual mean temperature°C40.48
Bio2Mean diurnal temperature range°C0.40
Bio3Isothermality (BIO2/BIO7) (×100)%23.75
Bio12Annual precipitationmm2.00
Bio13Precipitation in the wettest monthmm27.19
Bio14Precipitation in the driest monthmm5.85
1 Model contributions represent the average values of nine models established under the current and future climate change scenarios (SSP2-4.5 and SSP5-8.5) for five designated periods (1973–2000, 2021–2040, 2041–2060, 2061–2080, and 2081–2100).
Table
Table 2. Evaluation parameters of MaxEnt modeling of P. hysterophorus.
Table 2. Evaluation parameters of MaxEnt modeling of P. hysterophorus.
Evaluation ParameterBefore RarifyingAfter Rarifying
Species occurrence points16,3539234
AUC0.6120.776
TSS0.6240.788
Kappa0.5130.685
Table
Table 3. Change in the suitable habitat (%) of P. hysterophorus in different continents in future compared to the current climatic condition (1973–2000).
Table 3. Change in the suitable habitat (%) of P. hysterophorus in different continents in future compared to the current climatic condition (1973–2000).
Continent1973–2000
(km2)		SSP2-4.5 1SSP5-8.5 2
2021–20402041–20602061–20802081–21002021–20402041–20602061–20802081–2100
Africa4,063,351.5−0.28−16.597.928.751.110.1310.0410.94
Antarctica000000000
Asia2,225,5022.828.3328.2430.928.2711.8932.2938.12
Australia1,641,442.55.52−1.419.039.290.24−3.378.889.29
Europe463,756.5−13.5534.4953.1861.05−8.59−26.4071.7856.65
North America1,416,8521.580.9617.7320.14−0.290.6417.8622.05
Oceania55,3507.1310.8232.2134.30−0.78−7.7928.1526.76
South America3,309,093−0.89−6.055.155.681.801.835.245.66
1,2 Minus (−) sign indicates a decrease in the suitable habitat area of P. hysterophorus.
Table
Table 4. Predicted habitat suitability for P. hysterophorus in different nations in 2081–2100 under future climate scenarios (SSP2-4.5 and SSP5-8.5).
Table 4. Predicted habitat suitability for P. hysterophorus in different nations in 2081–2100 under future climate scenarios (SSP2-4.5 and SSP5-8.5).
Habitat Suitability 1SSP2-4.5SSP5-8.5
ModerateBrunei, Chile, China, Georgia, Iran, Japan, Liechtenstein, Montenegro, Niger, Syria, YemenBelgium, China, Georgia, Iran, Japan, Niger, Syria, Yemen
HighBelgium, Bulgaria, Djibouti, Slovenia, South KoreaBrunei, Bulgaria, Liechtenstein, Macedonia, Montenegro, Slovenia
Very highMonaco, Netherlands, Tokelau (New Zealand)Djibouti, Monaco, Netherlands, South Korea, Tokelau (New Zealand)
1 Moderate habitat suitability (0.25–0.50); high habitat suitability (0.51–0.75); and very high habitat suitability (0.76–1). Here, some countries with low habitat suitability (≤0.25) in 1973–2000 are predicted to transition into the moderate, high, and very high habitat suitability categories by 2081–2100 under the future climate scenarios SSP2-4.5 and SSP5-8.5. The habtiat suitability value of each category is presented in Table S7.
Table
Table 5. Estimation of suitable habitat for P. hysterophorus in the different ADs of South Korea under the current climate (1973–2000) and future climate scenarios (SSP2-4.5 and SSP5-8.5; 2021–2040, 2041–2060, 2061–2080, and 2081–2100).
Table 5. Estimation of suitable habitat for P. hysterophorus in the different ADs of South Korea under the current climate (1973–2000) and future climate scenarios (SSP2-4.5 and SSP5-8.5; 2021–2040, 2041–2060, 2061–2080, and 2081–2100).
ADTotal Area (km2)1973–2000SSP2-4.5SSP5-8.5
2021–20402041–20602061–20802081–21002021–20402041–20602061–20802081–2100
Busan193.500016218000184.5189
North Chungcheong192600011611453.5022.51624.51773
South Chungcheong21151845811264.514589922515931993.5
Daegu220.531.585.5108198207112.5130.5207216
Daejeon144004.51441441854144144
Gangwon43920013.5283.5418.500504886.5
Gwangju130.55463117130.5130.594.5117130.5130.5
Gyeonggi2695.5000193.5306003781057.5
North Gyeongsang4963.5193.533311523838.541409721381.54351.54599
South Gyeongsang27090181352178235814437823312524.5
Incheon261000000000
Jeju459004.5292.5310.500301.5337.5
North Jeolla2052220.536964817371813.5594796.518361944
South Jeolla3055.540.5157.550424842664306580.526372907
Sejong1170009911700117117
Seoul148.50000000036
Ulsan274.5013.5452342432727243256.5
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

Adhikari, P.; Lee, Y.-H.; Poudel, A.; Lee, G.; Hong, S.-H.; Park, Y.-S. Predicting the Impact of Climate Change on the Habitat Distribution of Parthenium hysterophorus around the World and in South Korea. Biology 2023, 12, 84. https://doi.org/10.3390/biology12010084

AMA Style

Adhikari P, Lee Y-H, Poudel A, Lee G, Hong S-H, Park Y-S. Predicting the Impact of Climate Change on the Habitat Distribution of Parthenium hysterophorus around the World and in South Korea. Biology. 2023; 12(1):84. https://doi.org/10.3390/biology12010084

Chicago/Turabian Style

Adhikari, Pradeep, Yong-Ho Lee, Anil Poudel, Gaeun Lee, Sun-Hee Hong, and Yong-Soon Park. 2023. "Predicting the Impact of Climate Change on the Habitat Distribution of Parthenium hysterophorus around the World and in South Korea" Biology 12, no. 1: 84. https://doi.org/10.3390/biology12010084

APA Style

Adhikari, P., Lee, Y. -H., Poudel, A., Lee, G., Hong, S. -H., & Park, Y. -S. (2023). Predicting the Impact of Climate Change on the Habitat Distribution of Parthenium hysterophorus around the World and in South Korea. Biology, 12(1), 84. https://doi.org/10.3390/biology12010084

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