Received: 18 November 2021 Accepted: 24 March 2022 DOI: 10.1111/afe.12502 OR I G I N A L A R T I C L E Model-based prediction of the potential geographical distribution of the invasive coconut mite, Aceria guerreronis Keifer (Acari: Eriophyidae) based on MaxEnt Owusu Fordjour Aidoo1,2 | Ricardo Siqueira da Silva3 | Paulo Antônio Santana Junior4 | Philipe Guilherme Corcino Souza3 | Rosina Kyerematen5 | Felix Owusu-Bremang6 | Ndede Yankey6 | Christian Borgemeister7 1Department of Biological, Physical and Mathematical Sciences, School of Natural and Environmental Sciences, University of Environment and Sustainable Development (UESD), Somanya, Ghana 2Institute of Teacher Education and Continuing Professional Development (ITECPD), University of Education (UEW), Winneba, Ghana 3Department of Agronomy, Universidade Federal dos Vales do Jequitinhonha e Mucuri (UFVJM), Diamantina, Brazil 4Department of Entomology, Universidade Federal de Viçosa, Viçosa, Brazil 5Department of Animal Biology and Conservation Sciences (DABCS), University of Ghana, Legon - Accra, Ghana 6Council for Scientific and Industrial Research (CSIR), Oil Palm Research Institute, Coconut Research Programme, Sekondi, Ghana 7Centre for Development Research (ZEF), University of Bonn, Bonn, Germany Correspondence Abstract Owusu Fordjour Aidoo, Department of Biological, Physical and Mathematical 1. The coconut mite Aceria guerreronis Keifer (Acari: Eriophyidae), is a destructive mite Sciences, School of Natural and Environmental pest of coconut, causing significant economic losses. However, an effective pest man- Sciences, University of Environment and Sustainable Development (UESD), PMB, agement strategy requires a clear understanding of the geographical areas at risk of the Somanya, E/R, Ghana. target pest. Email: ofaidoo@uesd.edu.gh 2. Therefore, we predicted the potential global distribution A. guerreronis using a Funding information machine learning algorithm based on maximum entropy. National Council for Scientific and 3. The potential future distribution for A. guerreronis covered the 2040 and 2060 periods Technological Development; the Universidade Federal dos Vales do Jequitinhonha e Mucuri under two climate change emission scenarios (SSP1-2.6 and SSP5-8.5) in the context of (UFVJM); the Minas Gerais State Research the sixth assessment report (AR6) of the Intergovernmental Panel on Climate Change. Foundation (Fundação de Amparo a Pesquisa do Estado de Minas Gerais – FAPEMIG); the 4. The MaxEnt model predicts the habitat suitability for A. guerreronis outside its pre- Brazilian Federal Agency for the Support and sent distribution, with suitable habitats in Oceania, Asia, Africa, and the Americas. Evaluation of Graduate Education (Coordenação de Aperfeiçoamento de Pessoal The habitat suitability for the pest will decrease from 2040 to 2060. de Ensino Superior – CAPES) 5. The areas with the highest risk of A. guerreronis are those with an annual average [Correction added on 28 April 2022, after first temperature of around 25 C, mean annual precipitation of about 1459 mm, mean online publication: SSP5-5.85 has been precipitation seasonality close to 64%, an average variation of daytime temperature corrected to SSP5-8.5 in point 3 of the  Abstract.] of about 8.6 C, and mean seasonality of temperature of about 149.7 C. 6. Our findings provide information for quarantine measures and policymaking, espe- cially where A. guerreronis is presently still absent. K E YWORD S Aceria guerreronis, climate change, coconut mite, machine-learning algorithm, MaxEnt Agr Forest Entomol. 2022;1–15. wileyonlinelibrary.com/journal/afe © 2022 Royal Entomological Society. 1 2 AIDOO ET AL. INTRODUCTION Significant economic copra losses associated with mite infestations ranged from 10% to 16% in West Africa (Julia & Mariau, 1979), The coconut palm (Cocos nucifera L.), is one of the most important 0%–30% in Mexico (Hernandez, 1977), and 0%–31.5% in St. Lucia commercial palms produced in the tropical and sub-tropical (Moore et al., 1989). regions, with coordinates ranging from 25 North to 25 South of Climate change plays an important role in the distribution of the equator (Ahuja et al., 2014). Coconut is thought to originate in species leading to either expansion or reduction of suitable habitats the region between Southwest Asia and Melanesia, where it (Aidoo et al., 2019; Ajene et al., 2020; Sung et al., 2018; Thomson shows the greatest genetic diversity based on modern molecular et al., 2010; Velásquez-Tibatá et al., 2013). Global warming is studies (Gunn et al., 2011). Coconut is a versatile crop that plays a expected to change the distribution of species around the world, by significant role in the social and economic lives of many rural com- changing their natural habitats (Velásquez-Tibatá et al., 2013). munities (Gurr et al., 2016). In Sri Lanka, coconut production Greenhouse gas emissions due to anthropogenic activities are asso- accounts for 2.5% of the country’s gross domestic product, 2.5% ciated with global warming (Ge et al., 2015), which may force cer- of export earnings, and 5% of jobs (L. C. P. Fernando et al., 2007). tain species to niche changes rather than dispersal to avoid It also provides fuel, food, shelter, and beverages (Foale, 2005). extinction (Román-Palacios & Wiens, 2020). Increasing CO2 levels The health benefits of coconuts are well documented, especially can have a significant impact on plants and pathogens. For instance, in providing calcium, magnesium, and potassium (Ahuja it can lead to an increase in the production of fungal spores and et al., 2014). thus the growth of pathogens, but can also increase plant resis- An invasive species, plant or animal that has been introduced tance to pathogens (Coakey et al., 1999), affecting existing pest outside its native range can pose a threat to the new area. Such spe- control strategies. Recent studies have focused on the effects of cies can be significant contributors to habitat loss, environmental climate change on crops (Mall et al., 2017), pests, and diseases depletion, and impairment, and their increasing spread is fueled both (Boland et al., 2004; Pautasso et al., 2012; Reddy, 2013), and the by global trade (Bradley et al., 2012; Westphal et al., 2008) as well as emergence of pest and disease outbreaks. Also, research has climate change (Pyšek & Richardson, 2010). In China, economic assessed the impact of climate change on abundance, distribution, losses due to invasive species between 2001 and 2003, were US$ ecological niche potential for arthropod pests, and the associated 14.45 billion (Xu et al., 2006), US$ 0.9–1.1 billion in Africa (Pratt losses in crop production and their impact on food security et al., 2017), and in the USA and Australia, annual losses are around (Gregory et al., 2005; Sharma & Prabhakar, 2014; Spathelf US$ 120 billion and 13.6, respectively (Hoffmann & et al., 2014). Broadhurst, 2016; Pimentel et al., 2005). However, invasive species Species distribution models (SDMs) employ computer algorithms are not only a major threat to the economies of the affected coun- to predict the habitat suitability of a species across geographical space tries, but also to agriculture, food security, and biodiversity conser- and time based on environmental data. Such models are important vation (Campbell et al., 2016; S. Kumar et al., 2014). Over the past tools to evaluate the potential response of pests to climate change 100 years, there has been a dramatic increase in the spread of inva- and have been useful at large spatial and temporal scales. For sive species due to increased trade, tourism, transportation, and instance, SDMs have been successfully used to model the habitat suit- travel (Hulme, 2009). ability of the cassava green mite Mononychellus tanajoa (Bondar) and The coconut mite Aceria guerreronis Keifa (Acari: Eriophyidae), Mononychellus mcgregori (Flechtmann and Baker) (both Acari: is an invasive pest that poses a serious threat to the coconut indus- Tetranychidae) (Parsa et al., 2015), tomato red spider mite Tetranychus try worldwide (Navia et al., 2013). The mite originated in the Ame- evansi (Baker & Pritchard) (Acari: Tetranychidae) (Meynard ricas but has been introduced into Asia and Africa (Návia et al., 2013), and palm mite Raoiella indica Hirst (Prostigmata: et al., 2005). The mite usually starts feeding on the fruits as early as Tenuipalpidae) (Amaro & de Morais, 2013). Although there are several 3 months after fertilization, but generally, the highest infestations SDM algorithm methods, Maximum Entropy (MaxEnt) employs are seen in 3–7 months old coconut fruits (L. C. P. Fernando presence-only data to predict the distribution of species based on the et al., 2003). Especially very young fruits, aged 1–4 months, are theory of maximum entropy (Phillips et al., 2005, 2006, 2017). It is highly susceptible to mite infestation (Howard & Rodriguez, 1991). generally considered the most powerful tool when modelling pests Mite infestation can affect the export acceptability of coconut with narrow ranges and where presence-only data exist (Baloch fruits (Rezende et al., 2016). Coconut mite can be detected by a vis- et al., 2020; Elith et al., 2006; S. Kumar et al., 2014; Y. Liu & ible triangular white spot on the fertilized nuts (Mariau, 1969), Shi, 2020; Phillips et al., 2006). found along the edge of the coconut’s perianth. Mite infestations Attempts to reduce the adverse effects of warming on insect cause reductions in the yield of copra, a premature drop of coconut, pests must provide a deeper understanding of the response of individ- reduced coconut fibre length and tensile strength, small deformed ual species and the complex ecological processes underlying their fruits, and result in a reduction in husk availability (Alam & response (Lehmann et al., 2020). Although the effect of climate Islam, 2014; Aratchige et al., 2016; P. P. Kumar & Ramaraju, 2010; change on pests of horticultural crops has considerably increased, Naseema Beevi et al., 2003; Wickramananda et al., 2007). information on its impact on the distribution of coconut mite, POTENTIAL DISTRIBUTION OF THE COCONUT MITE 3 essential for mitigation measures, is generally lacking. Therefore, for without it to check whether the adjustment will affect the habitat the first time, we used MaxEnt to model and map regions suitable for suitability predictions for A. guerreronis (Figure 1). coconut mites to provide information for the implementation of quar- antine measures and highlight the need for policy formulation. Environmental datasets MATERIALS AND METHODS The climatic data used for this study were gathered from the WorldClim environmental database (http://www.worldclim.org/). These Occurrence datasets environmental variables have been widely used for species distribution modelling (Fick & Hijmans, 2017; Hijmans et al., 2005). The variables In the present study, we conducted a nationwide survey from April are mean climate values ranging from 1970 to 2000, with a high resolu- 2019 to February 2021 in Ghana to obtain A. guerreronis occurrence tion of 2.5 arc-min (≈5 km at the equator) (Ramos et al., 2018). The records. The locations where A. guerreronis was found were datasets consisting of 19 bioclimatic variables are presented in Table 1. georeferenced using a handheld GPS device. The presence of the SDMtools in ArcGIS software are proven to be effective in withdraw- pest was based on the availability of any of the developmental ing highly correlated variables, maintaining only one variable per group stages and symptomatic fruits. These presence records were sup- with high correlation. Therefore, for the collection of variables, Pearson’s plemented by an extensive scientific literature search (Aratchige correlation coefficients were calculated (Table S1), and those variables et al., 2016; L. C. P. Fernando et al., 2003; Howard et al., 1990; with values greater/or equal to 0.7 were considered as strong correlation Lawson-Balagbo et al., 2008; Navia et al., 2013), utilizing online (S. Kumar et al., 2014; Rank et al., 2020). Thus, the variables that were databases, such as Web of Science, Science Direct, Google, Google used in the final model as predictor variables are highlighted in bold in Scholar, PubMed, and MEDLINE. We searched online using key- Table 1. The predictive contribution of environmental variables was esti- words, such as A. guerreronis, coconut mite, coconut eriophyid mites, mated using the jackknife technique (Wei et al., 2018; Yang et al., 2013). first report, coconut pest, distribution of coconut mite, and the biol- ogy of coconut mite. In addition, we also consulted databases such as Global Biodiversity Information Facility (GBIF; http://www.gbif. Model development and validation org) and the Centre for Agriculture and Bioscience International (CABI; https://www.cabi.org/). In case only localities were provided, The model for A. guerreronis was based on adjusting the default Google Earth Pro was used to extract the coordinates (i.e., latitudes MaxEnt settings for certain combinations of resource types and the and longitudes). Duplicate and fuzzy records were manually removed regularization multiplier (RM) (Jarnevich et al., 2015; Merow from the data. et al., 2013). We combined linear (L), quadratic (Q), product (P), threshold (T), and hinge (H) feature sets, using automatic feature selection plus RM = 1 to control the number of parameters and, Filtering of the A. guerreronis datasets therefore, the complexity of the model for the species (Elith et al., 2011). The area under the the curve (AUC) of the receiver Occurrence records for running the Maxent model are often biased in operating characteristic curve (ROC) value was used to evaluate the the geographical space because of unequal sampling efforts across model’s performance (Peterson et al., 2008). The AUC value of 0.5, the study area (Botella et al., 2020; Stolar & Nielsen, 2015). Moreover, indicates that model predictions are no better than random a distribution location is typically biased towards areas that are easily (Peterson et al., 2011). The values below 0.5 are less than random, accessible to humans, such as nearby cities and other human habita- whereas those between 0.5 and 0.7 indicate poor performance. AUC tion areas (Hijmans et al., 2005; Kadmon et al., 2004). This, however, values between 0.7 and 0.9 indicate moderate performance, while suggests that distribution site data can significantly influence the out- values greater than 0.9 indicate high performance. MaxEnt is a come of a model due to spatial autocorrelation (Elith et al., 2011; machine-learning method that predicts probability distribution based Raghavan et al., 2019). To reduce this spatial autocorrelation, we used on the principle of maximum entropy. The model generates a suit- the R software package spThin to perform spatial filtering on the data ability index ranging from 0 to 1, with 0 indicating inadequacy and (Aiello-Lammens et al., 2015). After this process, all remaining occur- 1 optimal suitability (Elith et al., 2011). We used ArcGIS version 10.8 rence points were at least 5 km apart (Boria et al., 2014; Veloz, 2009). to prepare the MaxEnt outputs and generate suitable and This distance was selected to restrict each cell to only one occurrence unsuitable areas for A. guerreronis. The Maximum Test Sensitivity point since we used 5-km spatial resolution climatic data in the Plus Specificity (MTSPS) threshold is considered simple and effective model. Therefore, leaving 137 localities for the final model. The data (C. Liu et al., 2005). Therefore, we used MTSPS to extract from the were then converted into MaxEnt-compatible formats prior to analy- predictive models, the unsuitable and suitable areas for sis. In addition, we run the model using only the Ghana data and A. guerrenoris. 4 AIDOO ET AL. F I GU R E 1 Current global distribution of Aceria guerreronis (a), the highlight of points removed after thinning (b); 137 remaining occurrence points used for the modelling (c) RESULTS exceeding a random distribution. During modelling, 80% of the occurring sites were randomly selected as training data and Modelling performance the other 20% were used to test the resulting models. To predict future areas, we used four suitability classes Based on the 19 bioclimatic datasets, the MaxEnt model pro- (unsuitable: 0-MTSPS; low: MTSPS-0.5; medium: 0.5–0.7 and vided satisfactory results with an AUC value of 0.975, thus, high: 0.7–1.0). POTENTIAL DISTRIBUTION OF THE COCONUT MITE 5 T AB L E 1 Nineteen environmental variables used for the model The environmental variable that decreased the gain the most when with code and units omitted was the average variation of daytime temperature (bio_2), Code Environmental variable Unit therefore, having the most information that is not present in the other variables (Figure 2a,b). bio_1 Annual average temperature C  MaxEnt generates response curves and the variables that hadbio_2 Average variation of day time temperature C  such a small effect in the model make them unlikely to be biologicallybio_3 Isothermality C meaningful. The results of the individual response curves for the dif- bio_4 Seasonality of temperature (SD  100) C ferent bioclimatic variables showed that the predicted probability of bio_5 Highest temperature of the hottest month C occurrence of A. guerreronis positively correlated with temperature bio_6 Lowest temperature of the coldest month C seasonality, annual precipitation and annual average temperature bio_7 Annual temperature variation C (Figure 3). The probability of A. guerreronis occurrence increased bio_8 Average temperature of the rainy quarter C sharply up to a value of 200C for temperature seasonality and months started declining afterwards (bio_4; Figure 3a). The probability of bio_9 Average temperature of the driest quarter C A. guerreronis presence was consistently steady until 10C when it months started increasing with the annual average temperature (bio_12; bio_10 Average temperature of the hottest quarter C Figure 3b). With increasing annual precipitation, the probability of months A. guerreronis presence increased steadily up to a value of 800 mm bio_11 Average temperature of the coldest quarter C before it started declining (bio_12; Figure 3c). months Globally, the results showed that the distribution of A. guerreronis bio_12 Annual precipitation mm occurs in areas with a mean seasonality of temperature of about bio_13 Precipitation of the rainiest month mm 149.67C, precipitation seasonality of about 64.62%, average variation bio_14 Precipitation of the driest month mm of daytime temperature close to 8.62C, annual precipitation close to bio_15 Precipitation seasonality % 1459 mm, and an annual average temperature around 25C (Table 2). bio_16 Precipitation of the rainiest quarter months mm The environmental variables that contributed most to the model were bi0_17 Precipitation of the driest quarter months mm seasonality of temperature (43.28%), annual average temperature bio_18 Precipitation of the hottest quarter months mm (22.35%), average variation of daytime temperature (18.98%), and bio_19 Precipitation of the coldest quarter months mm annual precipitation (12.86%). The thermal conditions combined contributed 84.61% to the model, whereas rainfall conditions 15.38%. Note: The environmental variables highlighted in bold were used for the final model. Predicted current distribution for A. guerreronis Contribution of environmental variables In the present study, A. guerreronis occurrence records were obtained The relative importance of environmental variables based on the jack- from Africa, Asia, and the Americas (Figure 4). The model predicts that knife test indicated that seasonality of temperature (bio_4) and annual highly suitable areas for A. guerreronis are primarily centered on the west average temperature (bio_1) contributed most to the model when and east coasts of Africa, the south and east coasts of Asia, the north used in isolation (Table 2). The environmental variable with the coast of Oceania, the east coast of South America, and the south coast of highest gain, thus providing the most useful information by itself North America (Figure 5). Moreover, the prediction agrees with the pre- when used in isolation, was the seasonality of temperature (bio_4). sent and proven records of A. guerreronis except for minor changes. T AB L E 2 Environmental variables considered in an Aceria guerreronis niche models and mean percentage contribution of environmental variables in an A. guerreronis distribution model; values were averaged over 10 repeated runs Percentage Permutation Variable contribution importance Min. Max. Avg. SD Annual average temperature (bio_1; C) 22.35 70.12 14.18 27.89 25.09 1.76 Annual precipitation (bio_12; mm) 12.86 2.47 85 2522 1459.19 666.61 Precipitation seasonality (CV) (bio_15; %) 2.52 1.75 19.28 112.91 64.62 26.74 Average variation of daytime temperature (bio_2; C) 18.98 16.09 4.60 12.62 8.62 1.67 Seasonality of temperature (SD  100) (bio_4; C) 43.28 9.57 22.60 375.73 149.67 92.58 Note: General statistics were calculated using all occurrences (n = 166). Abbreviations: Avg., average; max., maximum; min., minimum; SD, standard deviation. 6 AIDOO ET AL. F I GU R E 2 The relative importance of environmental variables based on (a) jackknife test of regularized training gain, and (b) jackknife of AUC for Aceria guerreronis F I GU R E 3 Response curves of the best predictors of Aceria guerreronis occurrence: (a) temperature seasonality (bio_4), (b) annual average temperature (bio_1) and (c) annual precipitation (bio_12). Red curves represent the average response and blue margins are1 SD computed over 10 replicates POTENTIAL DISTRIBUTION OF THE COCONUT MITE 7 F I GU R E 4 Present known distribution, and predicted suitable and unsuitable areas for Aceria guerreronis F I GU R E 5 Current time global Aceria guerreronis suitability classes Predicted future distribution for A. guerreronis predicts a reduction in habitat suitability from 2040 to 2060. How- ever, the areas that will remain medium to high suitability for The future predictions showed an expansion of habitat suitability A. guerreronis are mainly found on the west and east coasts of Africa, from the current time to the future. The model predicts that the east coast of South America, south and east coasts of Asia, and major coconut-producing countries, such as Indonesia, India, north coast of Oceania. Under SSP8.5 (Figure 7), the model predicts Sri Lanka, Brazil, Vietnam, and the Philippines, will continue to that suitable areas for A. guerreronis will decrease from 2040 to remain suitable in the future. Under SSP2.6 (Figure 6), the model 2060. However, habitat suitability for A. guerreronis will range from 8 AIDOO ET AL. F I GU R E 6 Predicted future distribution of Aceria guerreronis in the shared socio-economic pathway (SSP2.6) of climate conditions in 2040 and 2060 low to high. But, the areas with medium to high suitability are essen- occurrences (Figure 8b), and using only Ghana occurrences showed tially centered on the east and west coasts of Africa, south and east varying predictions (Figure 8c). The model’s prediction shows a slight coasts of Asia, east coast of South America, and north coast of expansion of habitat suitability for the pest when Ghana occurrences Australia. were removed from the datasets (Figure 8b). However, the model predicted suitable areas where the pest was identified and sampled. But, when only Ghana records were used in the model, the model Predicted current distribution for A. guerreronis using predicts suitable areas basically in Africa’s west and east coasts, only Ghana data and without it parts of the east and north coasts of South America, and the south- east coast of Asia (Figure 8c). The prediction with only Ghana data The current time global habitat suitability predictions for was also consistent with the known occurrence records of the pest A. guerreronis using all occurrences (Figure 8a), without Ghana in Ghana. POTENTIAL DISTRIBUTION OF THE COCONUT MITE 9 F I GU R E 7 Predicted future distribution of Aceria guerreronis in the shared socio-economic pathway (SSP8.5) of climate conditions in 2040 and 2060 DISCUSSION its high predictive power and, does not require absence records of the targeted species (Phillips et al., 2017). Previous studies have used the The invasive coconut mite is considered an economically important method to successfully predict climate suitable areas of several spe- pest of coconut wherever it occurs. Although its exact origin is cies on a local and global scale (S. Kumar et al., 2014; Ning unknown, it is most likely a South American native travelled with the et al., 2017; Santana Jr et al., 2019; Aidoo et al., 2022). Our model seeds to Africa and Asia (Návia et al., 2005). Therefore, the global performance was high, with an AUC value of more than 0.9, potential distribution maps of its expansion are essential for economic suggesting that the model’s predictions are reliable. Several studies and quarantine purposes worldwide, especially for areas where the have assessed MaxEnt model performance using AUC (Remya pest has not been reported. In the present study, we used the MaxEnt et al., 2015; Yang et al., 2013). model to predict the current and future distributions of A. guerreronis. In our model, the environmental variables that had the strongest MaxEnt theory has several advantages over other SDMs because of influence on coconut mite habitat suitability in descending order of 10 AIDOO ET AL. F I GU R E 8 Exploring the effect of the data from Ghana importance were seasonality of temperature, annual average tempera- distribution and spread of the coconut mite. These findings are consis- ture, average variation of daytime temperature, annual precipitation, tent with previous studies which indicated that temperature and precipitation seasonality. This suggests that thermal parameters influenced the seasonal abundance of A. guerreronis (Lawson-Balagbo much more than rainfall are the main factors driving the potential et al., 2008). This, however, confirms that the main distribution zones POTENTIAL DISTRIBUTION OF THE COCONUT MITE 11 are in the tropical and sub-tropical regions (Lawson-Balagbo largely facilitate making informed decisions on mitigation measures for et al., 2008). Earlier work has shown that the ideal temperature range researchers, agriculturists, ecologists, and other stakeholders. for A. guerreronis development is between 9.3 and 33.6C (Ansaloni & In the present study, we used bioclimatic variables and elevation Perring, 2002), but it can live for at least 5 h of frost and for over a datasets. But, there are several other biotic and/or abiotic factors, week below 5C (Howard et al., 1990; Navia et al., 2013). However, which were not considered in the present study. These factors may high temperatures above 40 are detrimental to the development of affect the biological invasion of A. guerreronis in the areas predicted to the pest (Navia et al., 2013). Fernando and Aratchige (2010), reported be suitable for the pest. For instance, host plants play a key role in the that damage levels and rate of spread are higher in the dry- and inter- distribution and establishment of invasive species (Lu et al., 2013), but mediate than in wet climates. In addition, long periods of drought and was not included in our model. Therefore, A. guerreronis host plants, high temperatures are significant climatic conditions influencing the such as coconut, Lytocaryum weddellianum (H. Wendl.), and Syagrus distribution and establishment of A. guerreronis (Lawson-Balagbo romanzoffiana (Cham.) (Návia et al., 2005), should be considered in et al., 2008). future predictions. Anthropogenic activities, including the transport of Our model predicts that climate suitable areas for A. guerreronis, coconut planting materials and international trade, may facilitate the are primarily centered on the west and east coasts of Africa, south spread of A. guerreronis to new areas (Sarkar, 2011). Finally, other fac- coasts of Asia, north and east coasts of South America, and coastal tors, including geographic barriers, land-use changes, conflict and regions of Northern Oceania. Our findings show an expansion of habi- reconstruction, regulatory regimes, tourism, public health factors, and tat suitability outside the known distribution of the pest, particularly environmental concerns may also influence the outcome of our pre- in Indonesia, Vietnam, Papua New Guinea, Mexico, Thailand, and dictions (FAO, 2001; McNeely et al., 2001; Perrings et al., 2002; Myanmar. In addition, the model’s predictions are consistent with the Pilcher, 2004; Zettler et al., 2004). These limitations should be consid- historical records of the pest (Návia et al., 2005). These predictions ered in future investigations. are quite disturbing because the major coconut-producing countries Using all occurrences, without Ghana occurrences, and using only where A. guerreronis has been reported will continue to be threatened Ghana occurrences data to run the models showed varying predictions. by the pest. Given that prevention of biological invasion is more eco- For instance, Ghana only data predicted habitat suitability for nomical than post-entry management (Angulo et al., 2021; Cuthbert A. guerreronis, mainly Ghana, Côte d’Ivoire, Togo, Benin, and Nigeria. et al., 2021; Fantle-Lepczyk et al., 2022), global efforts are required to This also supports previous studies that sample size affects predictions slow down the pest’s invasion. Previous work has shown that the of species habitat suitability (Støa et al., 2019; Syfert et al., 2013). It is most efficient strategy to prevent biological invasion is implementing worth mentioning that the results from the Ghana only data are a comprehensive quarantine program, and ecological evaluation and expected, given that the data cover so little of the global range. This monitoring programs (Xie et al., 2003). The presence of the pest in also shows how the model could lead to biases (Fourcade et al., 2014). Brazil and Sri Lanka may require physical, biological, and chemical con- A filtering method has been using in run models to reduce the spatial trol measures to either eradicate or control the pest outbreaks. How- aggregation of records (Aiello-Lammens et al., 2015; Boria et al., 2014; ever, our predictions will serve as a guide in developing a more Ramos et al., 2018; Santana Jr et al., 2019). However, all sample data efficient management program for the pest. are incomplete and potentially biased and may still have significant spa- In the present study, we used two socioeconomic pathways (SSPs tial autocorrelation when considering the global scale at which the 2.6 and 8.5). These pathways have been used to predict the impact of models are run (Jarnevich et al., 2015). Thus, it is helpful to run a model climate change on species (e.g., Kriegler et al., 2012). These are useful considering uneven distribution occurrences group. Therefore, the for determining the long-term consequences of human-induced climate interpretation of prediction considering the occurrences data should be change and provide useful information for preventing biological inva- taken into account to aim for precautionary measures to limit potential sion. For the SSPs, the model predicts a contraction of suitable areas introductions of A. guerreronis. The model should continually run when from SSPs 2.6 to 8.5. Rising global temperatures will mean that regions A. guerreronis records are updated in the future. that are at present unsuitable to coconut mites, especially in Brazil, India, Australia, Nigeria, and Kenya, will become suitable habitats from now until 2040. Yet, global warming will also cause the ecological range CONCLUSION of A. guerreronis to shift inwards, notably in parts of Kenya, Tanzania, Somalia, Cambodia, and Thailand. This will make areas that are pres- Climate change will have a significant impact on the global distribu- ently highly suitable less until 2060. However, it is important to note tion of A. guerreronis, based on the future model predictions from that the exact impact of climate change on invasive species like the current time to 2060, however, the specific effects will vary A. guerreronis may vary depending on the geographical location, the across different geographical locations. Our predictions will assist country’s wealth, and farm management practices (Nair et al., 2003; governments in making the most of their financial investments in Navia et al., 2013; Pratt et al., 2017). Moreover, the presence of natural pest control initiatives by determining areas that are or will become enemies, crop production practices, competition with native species, more or less suitable for current and potential future pest out- and lack of dispersal options, may prevent outbreaks in the invaded breaks. Therefore, we propose special precautionary measures to areas (Al-Shanfari et al., 2010). Notwithstanding, our findings will limit potential man-made introductions of A. guerreronis through 12 AIDOO ET AL. improved regulations of importation of coconuts or coconut prod- Aidoo, O.F., Tanga, C.M., Mohamed, S.A., Rasowo, B.A., Khamis, F.M., ucts into potentially suitable regions. Furthermore, research should Rwomushana, I. et al. (2019) Distribution, degree of damage and risk of spread of Trioza erytreae (Hemiptera: Triozidae) in Kenya. Journal focus on explorations and subsequent introductions and/or preser- of Applied Entomology, 143, 822–833. vation of indigenous natural enemies as part of comprehensive Aiello-Lammens, M.E., Boria, R.A., Radosavljevic, A., Vilela, B. & integrated pest management (IPM) strategy against the Anderson, R.P. (2015) spThin: an R package for spatial thinning of coconut mite. species occurrence records for use in ecological niche models. Ecography, 38, 541–545. Aidoo, O.F., Souza, P.G.C., Silva, R.S., Santana Júnior, P.A., Picanço, M.C., ACKNOWLEDGEMENTS Kyerematen, R. et al. (2022) Climate-induced Range Shifts of Inva- This research was supported by the National Council for Scientific and sive Species (Diaphorina citriKuwayama). Pest Management Science. Technological Development (Conselho Nacional de Desenvolvimento https://doi.org/10.1002/ps.6886  – Ajene, I.J., Khamis, F.M., van Asch, B., Pietersen, G., Seid, N.,Científico e Tecnologico CNPq), the Brazilian Federal Agency for the Rwomushana, I. et al. (2020) Distribution of Candidatus Liberibacter Support and Evaluation of Graduate Education (Coordenação de species in Eastern Africa, and the first report of Candidatus Aperfeiçoamento de Pessoal de Ensino Superior – CAPES) - Finance Liberibacter asiaticus in Kenya. Scientific Reports, 10, 1–10. Code 001, the Minas Gerais State Research Foundation (Fundação de Alam, N.S. & Islam, M.N. (2014) Mite management of coconut in Amparo a Pesquisa do Estado de Minas Gerais – FAPEMIG) and the Bangladesh. In Proceedings of the mite management workshop of coco- nut in SAARC member countries. SAARC Agriculture Centre (SAC), Universidade Federal dos Vales do Jequitinhonha e Mucuri (UFVJM) Dhaka, pp. 25–38. with the granting of resources to carry out this work. Al-Shanfari, A., Gowen, S.R., Moore, D., Hountondji, F.C.C. & de Moraes, G.J. (2010) The coconut mite Aceria guerreronis Keifer (Acari: CONFLICT OF INTEREST Eriophyidae) in the sub-arid coastal region of Dhofar, Oman. In XIII international congress of acarology. Abstracts. Recife (Vol. 11). The authors declare that they have no known competing financial Amaro, G. & de Morais, E.G.F. (2013) Potential geographical distribution of interests or personal relationships that could have appeared to influ- the red palm mite in South America. Experimental and Applied ence the work reported in this paper. Acarology, 60, 343–355. Angulo, E., Ballesteros-Mejia, L., Novoa, A., Duboscq-Carra, V., Diagne, C. & Courchamp, F. (2021) Economic costs of invasive alien AUTHOR CONTRIBUTIONS species in Spain. NeoBiota, 67, 267–297. Conceptualization, formal analysis, methodology, writing - original draft, Ansaloni, T., & Perring, T.M. (2002) Biology of Aceria guerreronis Keifer writing - review & editing, funding acquisition, project administration: on queen palms. In: Morales-Malacara JB, Rivas G (eds), XI interna- Owusu Fordjour Aidoo. Methodology, software, formal analysis, writing - tional congress of acarology. Program and Abstract Book. Universidad Nacional Auto'noma de Me'xico, Distrito Federal Me'xico, p. 145. review & editing: Ricardo Siqueira da Silva. Writing - review & editing, for- Aratchige, N.S., Kumara, A.D.N.T. & Suwandharathne, N.I. (2016) The mal analysis, writing - original draft, writing - review & editing: Paulo Antô- coconut mite: current global scenario. In: Economic and ecological sig- nio Santana Junior. Writing - original draft, Software, formal analysis, nificance of arthropods in diversified ecosystems. Singapore: Springer writing - review & editing, funding acquisition: Philipe Guilherme Corcino Nature, pp. 321–342. Baloch, M.N., Fan, J., Haseeb, M. & Zhang, R. (2020) Mapping potential Souza. Conceptualization, writing - original draft, writing - review & editing: distribution of Spodoptera frugiperda (Lepidoptera: Noctuidae) in cen- Rosina Kyerematen. Conceptualization, methodology, formal analysis, tral Asia. Insects, 11(3), 172. writing - original draft, writing - review & editing, funding acquisition: Felix Boland, G.J., Melzer, M.S., Hopkin, A., Higgins, V. & Nassuth, A. (2004) Cli- Owusu Bremang. Conceptualization, project administration, writing - mate change and plant diseases in Ontario. Canadian Journal of Plant review & editing: Ndede Yankey. Conceptualization, methodology, writing Pathology, 26, 335–350. Boria, R.A., Olson, L.E., Goodman, S.M. & Anderson, R.P. (2014) Spatial fil- - original draft, writing - review & editing: Christian Borgemeister. tering to reduce sampling bias can improve the performance of eco- logical niche models. Ecological Modelling, 275, 73–77. DATA AVAILABILITY STATEMENT Botella, C., Joly, A., Monestiez, P., Bonnet, P. & Munoz, F. (2020) Bias in The data that support the findings of this study are available from the presence-only niche models related to sampling effort and species niches: lessons for background point selection. PLoS One, 15, corresponding author upon reasonable request. e0232078. Bradley, B.A., Blumenthal, D.M., Early, R., Grosholz, E.D., Lawler, J.J., ORCID Miller, L.P. et al. (2012) Global change, global trade, and the next Owusu Fordjour Aidoo https://orcid.org/0000-0003-3332-0943 wave of plant invasions. Frontiers in Ecology and the Environment, 10, 20–28. Ricardo Siqueira da Silva https://orcid.org/0000-0003-1837-1339 Campbell, B.M., Vermeulen, S.J., Aggarwal, P.K., Corner-Dolloff, C., Philipe Guilherme Corcino Souza https://orcid.org/0000-0002- Girvetz, E., Loboguerrero, A.M. et al. (2016) Reducing risks to food 8117-2394 security from climate change. Global Food Security, 11, 34–43. Christian Borgemeister https://orcid.org/0000-0001-8067-0335 Coakley, S.M., Scherm, H. & Chakraborty, S. (1999) Climate change and plant disease management. Annual review of phytopathology, 37, 399–426. REFERENCES Cuthbert, R., Bartlett, A., Turbelin, A., Haubrock, P., Diagne, C., Pattison, Z. Ahuja, S.C., Ahuja, U. & Ahuja, S. (2014) Coconut-history, uses, and folk- et al. (2021) Economic costs of biological invasions in the lore. Asian Agri-History, 18(3), 221–248. United Kingdom. NeoBiota, 67, 299–328. POTENTIAL DISTRIBUTION OF THE COCONUT MITE 13 Elith, J., Graham, C.H., Anderson, R.P., Dudík, M., Ferrier, S., Guisan, A. Jarnevich, C.S., Stohlgren, T.J., Kumar, S., Morisette, J.T. & Holcombe, T.R. et al. (2006) Novel methods improve prediction of species’ distribu- (2015) Caveats for correlative species distribution modeling. Ecologi- tions from occurrence data. Ecography, 29, 129–151. cal Informatics, 29, 6–15. Elith, J., Phillips, S.J., Hastie, T., Dudík, M., Chee, Y.E. & Yates, C.J. (2011) A Julia, J.F. & Mariau, D. (1979) New research on the coconut mite, statistical explanation of MaxEnt for ecologists. Diversity and Distri- Eriophyes guerreronis K., in the Ivory Coast. Oleagineux, 34, butions, 17, 43–57. 181–189. Fantle-Lepczyk, J.E., Haubrock, P.J., Kramer, A.M., Cuthbert, R.N., Kadmon, R., Farber, O. & Danin, A. (2004) Effect of roadside bias on the Turbelin, A.J., Crystal-Ornelas, R. et al. (2022) Economic costs of bio- accuracy of predictive maps produced by bioclimatic models. Ecologi- logical invasions in the United States. Science of the Total Environ- cal Applications, 14, 401–413. ment, 806, 151318. Kriegler, E., O’Neill, B.C., Hallegatte, S., Kram, T., Lempert, R.J., Moss, R. FAO. (2001) The State of Food and Agriculture 2001. Rome. Available at: H. et al. (2012) The need for and use of socio-economic scenarios www.fao.org/docrep/003/x9800e/x9800e14.htm for climate change analysis: a new approach based on shared Fernando, L.C.P., Aratchige, N.S. & Peiris, T.S.G. (2003) Distribution pat- socio-economic pathways. Global Environmental Change, 22, terns of coconut mite, Aceria guerreronis, and its predator Neoseiulus 807–822. aff. paspalivorus in coconut palms. Experimental & Applied Acarology, Kumar, P.P. & Ramaraju, K. (2010) Impact of nut-infesting eriophyid mite, 31, 71–78. Aceria guerreronis Keifer, on the out-turn and quality of coconut fiber. Fernando, M.T.N., Zubair, L., Peiris, T.S.G., Ranasinghe, C.S. & Ratnasiri, J. In Proceedings of the XIII International Congress of Acarology. Abstracts. (2007) Economic value of climate variability impacts on coconut pro- Recife (Vol. 214). duction in Sri Lanka. Assessment of impact and adaptation to climate Kumar, S., Graham, J., West, A.M. & Evangelista, P.H. (2014) Using district- change, 45, 1–7. level occurrences in MaxEnt for predicting the invasion potential of Fick, S.E. & Hijmans, R.J. (2017) WorldClim 2: nouvelles surfaces an exotic insect pest in India. Computers and Electronics in Agriculture, climatiques de résolution spatiale de 1 km pour les zones ter- 103, 55–62. restres mondiales. International Journal of Climatology, 37, 4302– Lawson-Balagbo, L.M., Gondim, M.G.C., De Moraes, G.J., Hanna, R. & 4315. Schausberger, P. (2008) Exploration of the acarine fauna on coconut Foale, M. (2005) An introduction to the coconut palm. In: Batugal, P., palm in Brazil with emphasis on Aceria guerreronis (Acari: Eriophyidae) Ramanatha Rao, V. & Oliver, J. (Eds.) Coconut genetic resources. and its natural enemies. Bulletin of Entomological Research, 98, 83–96. Serdang: International Plant Genetic Resources Institute, Regional Lehmann, P., Ammunét, T., Barton, M., Battisti, A., Eigenbrode, S.D., Jepsen, J. Office for Asia, the Pacific and Oceania (IPGRI-APO). U. et al. (2020) Complex responses of global insect pests to climate Fourcade, Y., Engler, J.O., Rödder, D. & Secondi, J. (2014) Mapping species warming. Frontiers in Ecology and the Environment, 18(3), 141–150. distributions with MAXENT using a geographically biased sample of Liu, C., Berry, P.M., Dawson, T.P. & Pearson, R.G. (2005) Selecting thresh- presence data: a performance assessment of methods for correcting olds of occurrence in the prediction of species distributions. sampling bias. PLoS One, 9(5), e97122. Ecography, 28, 385–393. Ge, Q., Wang, H., Rutishauser, T. & Dai, J. (2015) Phenological response to Liu, Y. & Shi, J. (2020) Predicting the potential global geographical distribu- climate change in China: a meta-analysis. Global change biology, 21, tion of two Icerya species under climate change. Forests, 11, 684. 265–274. Lu, X., Siemann, E., Shao, X., Wei, H. & Ding, J. (2013) Climate warming Gregory, P.J., Ingram, J.S. & Brklacich, M. (2005) Climate change and food affects biological invasions by shifting interactions of plants and her- security. Philosophical Transactions of the Royal Society B: Biological bivores. Global Change Biology, 19, 2339–2347. Sciences, 360, 2139–2148. Mall, R.K., Gupta, A. & Sonkar, G. (2017) Effect of climate change on agri- Gunn, B.F., Baudouin, L. & Olsen, K.M. (2011) Independent origins of culti- cultural crops. In: Current developments in biotechnology and bioengi- vated coconut (Cocos nucifera L.) in the old world tropics. PLoS One, neering. Amsterdam: Elsevier, pp. 23–46. 6, e21143. Mariau, D. (1969) Aceria guerreronis Keifer: récent ravageur de la Gurr, G.M., Johnson, A.C., Ash, G.J., Wilson, B.A., Ero, M.M., Pilotti, C.A. cocoteraie dahoméenne. Oléagineux, 24, 269–272. et al. (2016) Coconut lethal yellowing diseases: a phytoplasma threat McNeely, J.A., Mooney, H.A., Neville, L.E., Schei, P. & Waage, J.K. (2001) A to palms of global economic and social significance. Frontiers in Plant global strategy on invasive alien species. Gland and Cambridge: World Science, 7, 1521. Conservation Union (IUCN). Hernandez, R.F. (1977) Combate químico de eriofido del cocotero Aceria Merow, C., Smith, M.J. & Silander, J.A., Jr. (2013) A practical guide to (Eriophyes) guerreronis (K) en la Costa Guerrero. –Agricultura técnica MaxEnt for modeling species’ distributions: what it does, and why en México, 4, 23–38. inputs and settings matter. Ecography, 36, 1058–1069. Hijmans, R.J., Cameron, S.E., Parra, J.L., Jones, P.G. & Jarvis, A. (2005) Very Meynard, C.N., Migeon, A. & Navajas, M. (2013) Uncertainties in high resolution interpolated climate surfaces for global land areas. predicting species distributions under climate change: a case study International Journal of Climatology: A Journal of the Royal Meteorologi- using Tetranychus evansi (Acari: Tetranychidae), a widespread agricul- cal Society, 25, 1965–1978. tural pest. PLoS One, 8(6), e66445. Hoffmann, B.D. & Broadhurst, L.M. (2016) The economic cost of managing Moore, D., Alexander, L. & Hall, R.A. (1989) The coconut mite, Eriophyes invasive species in Australia. NeoBiota, 31, 1–18. guerreronis Keifer in St. Lucia: yield losses and attempts to control it Howard, F.W., Abreu-Rodríguez, E. & Denmark, H.A. (1990) Geographi- with acaricide, polybutene and Hirsutella fungus. International Journal cal and seasonal distribution of the coconut mite, Aceria of Pest Management, 35(1), 83–89. guerreonis (Acari: Eriophyidae), in Puerto Rico and Florida, USA. Nair, C.P.R., Rajan, P. & Chandrika, M. (2003) Coconut eriophyid Mite The Journal of Agriculture of the University of Puerto Rico, 74, Aceria guerreronis Keifer. An overview. Indian Journal of Plant Protec- 237–251. tion, 33, 1–10. Howard, F.W. & Rodriguez, E.A. (1991) Tightness of the perianth of Naseema Beevi, S., Mathew, T.B., Bai, H. & Saradamma, K. (2003) Status cococuts in relation to infestation by coconut mites. The Florida Ento- of eriophyid mite in Kerala-Resume of work done. In Coconut erio- mologist, 74, 358–361. phyid mite: issues and strategies. Proceedings of the international work- Hulme, P.E. (2009) Trade, transport and trouble: managing invasive species shop on coconut mite. Bangalore, pp. 64–75. pathways in an era of globalization. Journal of Applied Ecology, 46, Navia, D., Gondim, M.G.C., Aratchige, N.S. & de Moraes, G.J. (2013) A 10–18. review of the status of the coconut mite, Aceria guerreronis (Acari: 14 AIDOO ET AL. Eriophyidae), a major tropical mite pest. Experimental and Applied Román-Palacios, C. & Wiens, J.J. (2020) Recent responses to climate Acarology, 59, 67–94. change reveal the drivers of species extinction and survival. Proceed- Návia, D., Moraes, G.J.D., Lofego, A.C. & Flechtmann, C.H. (2005) Acarofauna ings of the National Academy of Sciences of the United States of Amer- associada a frutos de coqueiro (Cocos nucifera L.) de algumas localidades ica, 117, 4211–4217. das Américas. Neotropical Entomology, 34, 349–354. Santana, P.A., Jr., Kumar, L., Da Silva, R.S., Pereira, J.L. & Picanço, M.C. Ning, S., Wei, J. & Feng, J. (2017) Predicting the current potential and (2019) Assessing the impact of climate change on the worldwide dis- future worldwide distribution of the onion maggot, Delia antiqua tribution of Dalbulus maidis (DeLong) using MaxEnt. Pest Manage- using maximum entropy ecological niche modeling. PLoS One, 12, ment Science, 75, 2706–2715. e0171190. Sarkar, P.K. (2011) Invasive coconut perianth mite, Aceria guerreroronis Keifer and Parsa, S., Hazzi, N.A., Chen, Q., Lu, F., Campo, B.V.H., Yaninek, J.S. et al. their sustainable management. Journal of Crop and Weed, 7, 184–190. (2015) Potential geographic distribution of two invasive cassava Sharma, H.C. & Prabhakar, C.S. (2014) Impact of climate change on pest green mites. Experimental and Applied Acarology, 65, 195–204. management and food security. In: Integrated pest management. Lon- Pautasso, M., Döring, T.F., Garbelotto, M., Pellis, L. & Jeger, M.J. (2012) don, UK: Academic Press, pp. 23–36. Impacts of climate change on plant diseases—opinions and trends. Spathelf, P., van der Maaten, E., van der Maaten-Theunissen, M., European Journal of Plant Pathology, 133, 295–313. Campioli, M. & Dobrowolska, D. (2014) Climate change impacts in Perrings, C., Williamson, M., Barbier, E.B., Delfino, D., Dalmazzone, S., European forests: the expert views of local observers. Annals of For- Shogren, J. et al. (2002) Biological invasion risks and the public good: est Science, 71, 131–137. an economic perspective. Conservation Ecology, 6, 1. Støa, B., Halvorsen, R., Stokland, J.N. & Gusarov, V.I. (2019) How much is Peterson, A.T., Papeş , M. & Soberon, J. (2008) Rethinking receiver operat- enough? Influence of number of presence observations on the per- ing characteristic analysis applications in ecological niche modeling. formance of species distribution models. Sommerfeltia, 39, 1–28. Ecological Modelling, 213, 63–72. Stolar, J. & Nielsen, S.E. (2015) Accounting for spatially biased sampling Peterson, A.T., Soberon, J., Pearson, R.G., Anderson, R.P., Martínez- effort in presence-only species distribution modelling. Diversity and Meyer, E., Nakamura, M. et al. (2011) Ecological niches and geographic Distributions, 21, 595–608. distributions (MPB-49). United States: Princeton University Press. Sung, S., Kwon, Y.S., Lee, D.K. & Cho, Y. (2018) Predicting the potential Phillips, S.J., Anderson, R.P., Dudík, M., Schapire, R.E. & Blair, M.E. (2017) distribution of an invasive species, Solenopsis invicta Buren Opening the black box: an open-source release of Maxent. (Hymenoptera: Formicidae), under climate change using species dis- Ecography, 40, 887–893. https://doi.org/10.1111/ecog.03049 tribution models. Entomological Research, 48, 505–513. Phillips, S.J., Anderson, R.P. & Schapire, R.E. (2006) Maximum entropy Syfert, M.M., Smith, M.J. & Coomes, D.A. (2013) The effects of sampling modeling of species geographic distributions. Ecological Modelling, bias and model complexity on the predictive performance of MaxEnt 190, 231–259. species distribution models. PLoS One, 8, e55158. Phillips, S.J., Dudík, M. & Schapire, R. E. (2005) Maxent software for species dis- Thomson, L.J., Macfadyen, S. & Hoffmann, A.A. (2010) Predicting the tribution modeling. Available at: www.cs.princeton.edu/schapire/maxent effects of climate change on natural enemies of agricultural pests. Pilcher, H.R. (2004) US Army may have killed Italian trees. Nature, Biological Control, 52, 296–306. 428, 713. Velásquez-Tibatá, J., Salaman, P. & Graham, C.H. (2013) Effects of climate Pimentel, D., Zuniga, R. & Morrison, D. (2005) Update on the environmen- change on species distribution, community structure, and conserva- tal and economic costs associated with alien-invasive species in the tion of birds in protected areas in Colombia. Regional Environmental United States. Ecological Economics, 52, 273–288. Change, 13, 235–248. Pratt, C.F., Constantine, K.L. & Murphy, S.T. (2017) Economic impacts of Veloz, S.D. (2009) Spatially autocorrelated sampling falsely inflates mea- invasive alien species on African smallholder livelihoods. Global Food sures of accuracy for presence-only niche models. Journal of Biogeog- Security, 14, 31–37. raphy, 36, 2290–2299. Pyšek, P. & Richardson, D.M. (2010) Invasive species, environmental Wei, B., Wang, R., Hou, K., Wang, X. & Wu, W. (2018) Predicting the cur- change and management, and health. Annual Review of Environment rent and future cultivation regions of Carthamus tinctorius L. using and Resources, 35, 25–55. MaxEnt model under climate change in China. Global Ecology and Raghavan, R.K., Barker, S.C., Cobos, M.E., Barker, D., Teo, E.J.M., Foley, D. Conservation, 16, e00477. H. et al. (2019) Potential spatial distribution of the newly introduced Westphal, M.I., Browne, M., MacKinnon, K. & Noble, I. (2008) The link long-horned tick, Haemaphysalis longicornis in North America. Scien- between international trade and the global distribution of invasive tific Reports, 9, 1–8. alien species. Biological Invasions, 10, 391–398. Ramos, R.S., Kumar, L., Shabani, F. & Picanço, M.C. (2018) Mapping global risk Wickramananda, R., Peiris, T.S.G., Neil Fernandol, M.T., Fernand, L.C. levels of Bemisia tabaci in areas of suitability for open field tomato culti- P. & Edgington, S. (2007) Impact of the coconut mite (Aceria vation under current and future climates. PLoS One, 13, e0198925. guerreronis Keifer) on the coconut industry in Sri Lanka. Cord, 23, Rank, A., Ramos, R.S., da Silva, R.S., Soares, J.R.S., Picanço, M.C. & 1–16. Fidelis, E.G. (2020) Risk of the introduction of Lobesia botrana in suit- Xie, Z., Chen, Z., Fan, D. & Xiong, G. (2003) Global consequences and con- able areas for Vitis vinifera. Journal of Pest Science, 93, 1167–1179. trol strategies of biological invasion. Ying Yong Sheng tai xue bao= Reddy, P.P. (2013) Impact of climate change on insect pests, pathogens and The Journal of Applied Ecology, 14, 1795–1798. nematodes. Pest Management in Horticultural Ecosystems, 19, 225–233. Xu, H., Ding, H., Li, M., Qiang, S., Guo, J., Han, Z. et al. (2006) The distribu- Remya, K., Ramachandran, A. & Jayakumar, S. (2015) Predicting the cur- tion and economic losses of alien species invasion to China. Biological rent and future suitable habitat distribution of Myristica dactyloides Invasions, 8, 1495–1500. Gaertn. Using MaxEnt model in the Eastern Ghats, India. Ecological Yang, X.Q., Kushwaha, S.P.S., Saran, S., Xu, J. & Roy, P.S. (2013) Maxent Engineering, 82, 184–188. modeling for predicting the potential distribution of medicinal plant, Rezende, D., Melo, J.W., Oliveira, J.E. & Gondim, M.G. (2016) Estimated Justicia adhatoda L. in Lesser Himalayan foothills. Ecological Engineer- crop loss due to coconut mite and financial analysis of controlling the ing, 51, 83–87. pest using the acaricide abamectin. Experimental and Applied Zettler, J.A., Taylor, M.D., Allen, C.R. & Spira, T.P. (2004) Consequences of Acarology, 69, 297–310. forest clear-cuts for native and nonindigenous ants (Hymenoptera: POTENTIAL DISTRIBUTION OF THE COCONUT MITE 15 Formicidae). Annals of the Entomological Society of America, 97, 513–518. How to cite this article: Aidoo, O.F., da Silva, R.S., Santana Junior, P.A., Souza, P.G.C., Kyerematen, R., Owusu-Bremang, SUPPORTING INFORMATION F. et al. (2022) Model-based prediction of the potential Additional supporting information may be found in the online version geographical distribution of the invasive coconut mite, Aceria of the article at the publisher’s website. guerreronis Keifer (Acari: Eriophyidae) based on MaxEnt. Agricultural and Forest Entomology, 1–15. Available from: https: Table S1. Cross-correlation (Pearson correlation coefficient, r) //doi.org/10.1111/afe.12502 between environmental variables