In this study, we identiﬁed the current distribution of ﬁve globally distributed invasive Hemidactylus species and predicted their potential and future distribution using species distribution models based on climate and elevation data. These species included H. brookii, H. frenatus, H. garnotii, H. mabouia,and H. turcicus. We show that many regions with tropical and Mediterranean climates are suitable for most of these species. However, their current and potential distributions sug- gest that climate is not the only limiting factor. We hypothesize that climatic conditions may affect competition and other interactions resulting in a segregated distribution of the studied Hemidactylus species. As an effect of global climate change it is likely that H. brookii will expand its range to areas that are currently colonized by H. mabouia and/or H. frenatus,while H. turcicus is likely to expand its range to areas that are not yet invaded by any Hemidactylus species. The role of species interactions in the range expansion of these ﬁve Hemidactylus species still remains poorly understood, but could be of major importance in understanding and managing these invasive species. Key words: Hemidactylus, invasive species, species distribution modeling, climate change Invasive species are a major cause of various environmental prob- Here, we investigate the distribution of five invasive gecko spe- lems ranging from biodiversity loss to disrupted ecosystem services cies from the genus Hemidactylus. Several species of house geckos (Gurevitch and Padilla 2004; Butchart et al. 2010). In the United from this genus belong to the world’s most widely distributed and States alone invasive species are estimated to be responsible for an invasive lizards. The Asian house gecko Hemidactylus frenatus, for economic burden of $120 billion annually related to, for example, example, is not including central and south America, northern pest control, crop damage, and loss of ecosystem services such as Australia as well as parts of Africa (Ro ¨ dder et al. 2008). Other spe- pollination (Pimentel et al. 2005; Cook et al. 2006). Invasive species cies such as the tropical house gecko Hemidactylus mabouia have are also considered to be a major threat to numerous threatened and also been introduced to many regions outside their native range endangered species (Gurevitch and Padilla 2004). Invasive herpeto- (Ro ¨ dder et al. 2008). The introduction of Hemidactylus species may fauna are an important cause of the global decline in biodiversity. lead to various problems to native lizard species through processes This becomes particularly evident when considering that species such as competition and the introduction of exotic parasites, ultim- richness of these invasive species is much higher in biodiversity hot- ately leading to the decline of native species (Petren and Case 1996; spots (Li et al. 2016). In addition, climate change may exacerbate Hoskin 2011). Another problem associated with Hemidactylus spe- these impacts because it affects both the establishment of new inva- cies is their fecal droppings, which can be a source of salmonella poi- sive species and the distribution of existing invasive species soning in people (Callaway et al. 2011). (Hellmann et al. 2008). However, these effects are species depend- Several Hemidactylus species are highly adapted to living in close ent; where some invasive species may thrive in a changing climate proximity to people. They oftentimes feed on insects that are at- others may not. Species distribution models (SDM) are a useful tool tracted to artificial light sources (Tkaczenko et al. 2014). Some to predict these effects by modeling the distribution of species under Hemidactylus species are also known to forage in garbage bins and current and future climatic conditions (Jeschke and Strayer 2008). on tables where they feed on leftovers such as boiled rice (Weterings V C The Author (2017). Published by Oxford University Press. 1 This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact Downloaded from https://academic.oup.com/cz/advance-article-abstract/doi/10.1093/cz/zox052/4101659 email@example.com by Ed 'DeepDyve' Gillespie user on 08 June 2018 2 Current Zoology, 2017, Vol. 00, No. 00 2017). These highly opportunistic behaviors have contributed to predict their potential and future distribution. We used historical oc- their successful invasion of many regions. Due to living in close currence data to map the current distribution and create species dis- proximity to people, they get accidentally stowed away in boxes tribution models in order to predict their potential and future which are placed on boats, cars or trucks, allowing them to quickly distribution. We discuss these results in the context of species inter- disperse over large distances (Heinsohn 2003; Short and Petren actions and how the distribution of one Hemidactylus species may 2011; Norval et al. 2012). Females of some species, such as, H. fre- affect the distribution of others. natus are able to store sperm for many months. Therefore, the spe- cies can reproduce easily in areas where it is newly introduced if environmental conditions are suitable (Murphy-Walker and Haley Materials and Methods 1996). Individuals of Hemidactylus species arrive in many areas of Species occurrence the world, but not always manage to establish viable populations. Species occurrence records for H. brookii (n ¼ 4,150), H. frenatus For example, Hemidactylus mabouia has been found as stow-aways 1 (n ¼ 19,609), H. garnotii (n ¼ 1,659), H. mabouia (n ¼ 7,672), and in several European countries such as the Netherlands and 2 H. turcicus (n ¼ 7,597) were retrieved from the Global Biodiversity Iceland . Climatic conditions at these locations are limiting the es- Information Facility (GBIF) database (2 January 2017; www.gbif. tablishment of populations from these species. However, under a org). This database contains species records from museum collec- changing climate these conditions may become more favorable and tions and other published sources. These GBIF data were considered certain species may be able to also establish in these regions. sufficient because of two reasons. First, the database contained a The current study focused on the five most common and invasive very large number of occurrences which very well approximates the Hemidactylus species; H. brookii Gray, 1845, H. frenatus Schlegel, known range of all five species. Second, the five focal species are 1836, H. garnotii Dume ´ ril and Bibron 1836, H. mabouia (Moreau often found in urban areas and residential situation and therefore de Jonne ` s, 1818), and H. turcicus (L, 1758). Hemidactylus brookii, easily observed. Thus, it is unlikely that areas where the species is H. frenatus, and H. garnotii are native to Asia and are genetically common are not included in the data. In addition, H. frenatus, for similar, belonging to the tropical Asian clade. Hemidactylus example, is mostly found within the urban areas and less so in the mabouia is native to Africa and belongs to the African Atlanctic natural habitat surrounding these urban areas (Newbery and clade and H. turcicus is native to the Mediterranean and belongs to Jones 2007). Therefore, we can safely assume that the occurrence re- the Arid clade (Carranza and Arnold 2006; Bauer et al. 2010). cords for Hemidactylus species in the GBIF database provide a suffi- Hemidactylus brookii, H. frenatus and H. garnotii are sympatric in cient base for distribution modelling. For the SDM, presence data their native range where H. frenatus is often the more dominant spe- were needed. Therefore, these records had to be cleaned and filtered cies of the three (Zug et al. 2007). Hemidactylus mabouia and in order to remove double records, records from the exact same lo- H. brookii are sympatric in large parts of H. mabouia’s native range cation, and erroneous geo-referenced records. (Leache ´ 2005). Hemidactylus frenatus, and H. mabouia are sympat- First all records were geocoded using the described localities pro- ric in their invasive range in parts of the Americas (Krysko et al. vided with each record. Records containing only country informa- 2003) as well as H. garnotii and H. turcicus (Meshaka Jr. 2000). tion were removed from the dataset. For all other records, the Certain interactions between these different species may cause shifts described localities were checked for inconsistencies and other fea- in abundance. Interactions may include competition for food or tures that might cause problems during the geocoding process. space resulting in fighting or sexual interference (Frankenberg 1984; Inconsistent records were corrected where possible and otherwise Dame and Petren 2006; Hoskin 2011). In certain parts of Florida, deleted from the dataset. The localities were then geocoded using H. turcicus was one of the most common invasive Hemidactylus spe- the Bing Maps API (Microsoft 2017). For all records that could not cies (Meshaka Jr 1995). After introduction of the much larger be correctly geocoded the localities were re-checked. At this stage, H. garnotii [snout vent length: 49–66 mm versus 40–60 mm the most common problem was the usage of old colonial names in (Selcer 1986; Zug et al. 2007)] this species became slowly displaced, historical records such as “Batavia, Indonesia” instead of “Jakarta, because H. garnotii has superior fighting abilities (Frankenberg Indonesia”. In these instances, the names were looked up using vari- 1984). In Florida and certain Pacific islands, however, H. garnotii ous internet resources and the current name was added to the record on its turn is being displaced by the smaller H. frenatus [SVL: 42–59 in order to re-geocode the given records. mm (Zug et al. 2007)] and H. mabouia [(SVL: 40–61 mm, Meshaka After all records were geocoded the dataset was minimized to Jr 1995; Dame and Petren 2006; Iturriaga and Marrero 2013]. contain only one record per location for each species. These records Displacement by H. frenatus is thought to be caused by sexual inter- were then overlaid with a map containing country boundaries using ference, where the sexual H. frenatus appears to interfere with the a geographical information system (ArcGIS 10.1; ESRI 2012). Based asexual H. garnotii (Dame and Petren 2006). Interactions as such on this overlay, we removed records where the country code from may have a strong effect on the distribution and population size of the dataset did not correspond with those from the map. The final these highly invasive species and are important to take into consider- dataset contained a total of 6,404 records for Hemidactylus ation when evaluating possibilities to control spreading of invasive brookii (n ¼ 720), H. frenatus (n ¼ 2,065), H. garnotii (n ¼ 359), species and develop management strategies. H. mabouia (n ¼ 1,429), and H. turcicus (n ¼ 1,831). In this study, we aimed to 1) identify the current distribution of five globally distributed invasive Hemidactylus species and 2) Species distribution modelling 1 Preserved specimen in the Natural History Museum Rotterdam. We developed SDMs using both MaxEnt and ensemble models. Specimen collected in 2001 in the Merwede Harbor, Rotterdam, MaxEnt stands for maximum entropy and is a modeling software Netherlands. specifically designed for SDMs (Phillips et al. 2006). MaxEnt allows 2 Preserved specimen in the Yale Peabody Museum if Natural History. the usage of presence-only data to model the probability of occur- Speciemen collected in 2009 in Iceland. rence based on a species’ environmental constraints. MaxEnt is a Downloaded from https://academic.oup.com/cz/advance-article-abstract/doi/10.1093/cz/zox052/4101659 by Ed 'DeepDyve' Gillespie user on 08 June 2018 Weterings and Vetter Distribution of invasive house geckos 3 Figure 1. Current distribution of ﬁve Hemidactylus species. Each dot displays a location where one or more specimen was recorded. Some solitary dots in north- ern regions, show specimen that were discovered during transportation in, for example, freight containers and boxes. These records were not excluded because they display the potential of the species to spread to other regions. Downloaded from https://academic.oup.com/cz/advance-article-abstract/doi/10.1093/cz/zox052/4101659 by Ed 'DeepDyve' Gillespie user on 08 June 2018 4 Current Zoology, 2017, Vol. 00, No. 00 Table 1. Co-occurrence of historical records for ﬁve hemidactylus species. Each row gives the percentage of occurrence records for a given species that co-occur with the species in each column. Where rows intersect with the column of the same species the percentage of occur- rence records that do not co-occur with other species is given. When a species occurrence record was found in 0.5 degree (55.5 km) grid cell with any of the other species it was considered to co-occur. H. brookii (%) H. frenatus (%) H. garnotii (%) H. mabouia (%) H. turcicus (%) H. brookii 47.1 19.1 4.5 25.7 3.6 H. frenatus 9.5 66.4 11.1 7.8 5.1 H. garnotii 10.4 52.2 11.4 14.4 11.4 H. mabouia 17.5 10.7 4.2 61.7 5.8 H. turcicus 3.2 9.1 4.4 7.6 75.8 Table 2. Results from the distribution models for all ﬁve species. The upper part of the table gives the test statistics in which AUC stands for the Area Under the Curve. The lower part of the table gives the importance (%) of each model parameter for each species. Species H. brookii H. frenatus H. garnotii H. mabouia H. turcicus MaxEnt Ensemble MaxEnt Ensemble MaxEnt Ensemble MaxEnt Ensemble MaxEnt Ensemble AUC test samples 0.902 0.954 0.885 0.964 0.966 0.990 0.902 0.965 0.904 0.974 Kappa 0.615 0.786 0.835 0.761 0.827 TSS 0.786 0.822 0.914 0.810 0.846 Number of training samples 612 1,756 306 1,215 1,557 Number of test samples 108 309 53 214 274 Permutation importance Mean temp. 9.4 11.5 7.4 29.3 8.1 19.9 12.2 38.1 53.2 32.1 Max temp. of hottest month 0.4 6.6 3.8 6.9 1.7 8.5 13.3 20.2 7.9 10.9 Min temp. of coldest month 76.3 57.4 57.5 33.2 29.5 18.9 47.8 24.2 16.0 14.9 Annual precipitation 5.2 8.7 10.1 2.4 19.7 8.7 9.1 4.9 8.5 15.3 Precipitation of driest month 2.9 5.2 0.5 1.1 4.8 6.3 3.3 2.9 1.6 6.6 Precipitation of hottest quarter 2.9 6.0 11.8 18.8 14.9 11.8 10.5 5.0 4.5 12.0 Elevation 2.9 4.6 8.9 8.2 21.3 26.0 3.6 4.6 8.2 8.2 very popular method for modeling presence-only data and has previ- climatic range for both training and testing the models. The output ously been used for modelling the distribution of Hemidactylus spe- of the models was maps showing the probability of occurrence for cies (Ro ¨ dder et al. 2008; Ro ¨ dder and Lo ¨ tters 2009; Kurita 2013). each species. Model fit was evaluated using the area under the curve (AUC) for both model types. The MaxEnt models were also assessed For the ensemble models, we used the biomod2 library (Thuiller by comparing the omission rate with the predicted omission. While et al. 2009) for R 3.4.0 (R Development Core Team 2017). for the ensemble models we also assessed the TSS and Kappa statis- Ensemble models combine various algorithms to calculate an aver- tics. An AUC value higher than 0.8 generally indicates a good model aged model (Thuiller et al. 2009). We used only algorithms that fit (Swets 1988), for the TSS and Kappa statistics values higher than allowed the usage of weights in order to incorporate sampling bias. 0.5 are generally considered good (Allouche et al. 2006). These algorithms included: Generalized Linear Models (GLM), An important and often overlooked issue with SDMs is sampling General Additive Models (GAM), Multiple Adaptive Regression bias (Fourcade et al. 2014). Species occurrence data may be clus- Splines (MARS), Generalized Boosting Model (GBM), Classification tered around certain areas, which may over-represent and/or under- Tree Analysis (CTA), Flexible Discriminant Analysis (FDA), and represent the species in a given locality and its environmental Artificial Neural Networks (ANN). Individual models were only characteristics. For example, in the vicinity of a long-term ecological included in the ensemble model when the AUC score was higher research station many samples may be collected over time while an than 0.8 and the TSS score higher than 0.7. adjacent area may not be sampled at all. These data might suggest Both MaxEnt and ensemble models in this study used the previ- that the species is more common around the research station while ously described presence-only records and 10,000 random back- in fact the species could be equally distributed in both areas. A sam- ground points to develop models for the current and future pling bias as such may be particularly evident in species that are probability of occurrence of the five invasive Hemidactylus species. abundant in urban areas, which are readily accessible and highly The background points were used in the model to sample the density populated. Such urban species may very well occur in natural habi- of covariates from the landscape (world) and compare these to those tats, but are simply sampled more often in urban areas. This sam- from the points where the species is known to be present (Elith et al. pling bias can be accounted for in MaxEnt using data that 2011). For both model types, we used 15% of randomly selected approximates the sampling bias. Within MaxEnt a bias file can be samples as test samples. Climatic niche shifts have been observed for included, in which higher values indicate a higher bias. For ensemble non-native herpetofauna, therefore using only a species’ native range models the sampling bias can be accounted for by using model for model fitting will result in a poor model (Li et al. 2014). Hence, weights in which records with higher weights are given more im- we did not differentiate between a species native and non-native portance in the model. We calculated two different sets of values range within the models and thus included a species’ complete Downloaded from https://academic.oup.com/cz/advance-article-abstract/doi/10.1093/cz/zox052/4101659 by Ed 'DeepDyve' Gillespie user on 08 June 2018 Weterings and Vetter Distribution of invasive house geckos 5 Table 3. Total potential area per species for which Maxent values are 50% or higher. Area is given in million km2. Species H. brookii H. frenatus H. garnotii H. mabouia H. turcicus Current MaxEnt 15.4 7.5 1.7 7.3 6.3 Ensemble 9.6 26.1 1.6 17.5 14.6 RCP 2.6 MaxEnt 22.5 (46%) 9.1 (22%) 1.4 (-17%) 4.3 (-41%) 6.8 (8%) Ensemble 15.2 (58%) 25.3 (-3) 1.1 (-30%) 9.8 (-44%) 16.0 (10%) RCP 4.5 MaxEnt 24.1 (56%) 9.0 (21%) 1.4 (-19%) 3.6 (-51%) 6.8 (8%) Ensemble 17.0 (76%) 26.0 (0%) 1.1 (-32%) 8.3 (-52%) 16.4 (12%) RCP 6.0 MaxEnt 23.8 (55%) 9.0 (20%) 1.4 (-19%) 3.7 (-49%) 6.8 (9%) Ensemble 16.3 (69%) 25.9 (-1%) 1.1 (-32%) 9.1 (-48%) 16.1 (10%) RCP 8.5 MaxEnt 28.5 (85%) 8.6 (14%) 1.2 (-27%) 3.3 (-55%) 8.0 (27%) Ensemble 18.8 (95%) 25.5 (-2%) 1.0 (-36%) 7.1 (-60%) 17.1 (17%) Figure 2. Comparison of probabilities of occurrence for the MaxEnt models versus the ensemble models for all ﬁve Hemidactylus species. Differences between models were analysed using the Mann-whitney U-test. P-values < 0.05 are indicated with * and P-values <0.001 are indicated with *** that approximated the sampling bias. First, we calculated the density for both the MaxEnt and ensemble model. This allowed us to assess of sampling points using a density kernel (Fourcade et al. 2014) and how suitable these location are for the specific species, whether second, we calculated the distance to urban areas and used this as species are likely to co-occur and to quantify overlap in certain en- our sampling bias. Urban areas were derived from global land cover vironmental conditions. Resulting values were compared using a data (1 km spatial resolution) that were retrieved from the United Mann–Whitney U-test and plotted in a series of boxplots using States Geological Survey (http://landcover.usgs.gov/documents/ RStudio (RStudio 2013) built on R 3.4.0 (Development Core Team GlobalLandCover_tif.zip), after which we calculated the distance to 2017). We also sampled the probability of occurrence in order to compare predicted values for both model types. All plots were cre- urban areas for each grid cell using the Euclidean Distance function in ArcGIS 10.1. These distance values were then inverted, giving ated using the ggplot2 package for R (Wickman 2009), while all urban areas the highest values and remote areas the lowest values. maps were created using ArcGIS 10.1. Using the bias correction based on the distance to urban areas re- sulted consistently in better models in comparison to the density ker- nel. Therefore, all MaxEnt models were fitted using the inverted Environmental data distance to urban areas for correcting the sampling bias. In the en- Environmental data that were used in the SDMs consisted of semble models, the sampling bias can be accounted for by including mapped so-called bioclim variables, which are available from http:// weights. Lower weights give an occurrence records less importance www.worldclim.org/. Bioclim variables consist of a set of 19 vari- in the model. Therefore, we did not use the inverted distance to ables that represent climatic variables that may be relevant to the urban areas but the actual distance. biological and ecological characteristics of certain species; these Finally, we sampled the predicted probability of occurrence of data are based on the climate data developed by Hijmans et al. one species on the location of all other species (Rodder et al. 2006), (2005). Six variables were selected based on 1) their ability to Downloaded from https://academic.oup.com/cz/advance-article-abstract/doi/10.1093/cz/zox052/4101659 by Ed 'DeepDyve' Gillespie user on 08 June 2018 6 Current Zoology, 2017, Vol. 00, No. 00 Figure 3. Results from the MaxEnt and ensemble models for H. brookii based on current climate conditions and projected for 2,050 using two representative con- centrations pathways (RCP). (A) Current potential distribution for the MaxEnt and (B) the ensemble model. (C) Predicted potential distribution under RCP 2.6 for the MaxEnt model and (D) the ensemble model. (E) Change in probability of occurrence between the current and future potential distribution under RCP 2.6 for the MaxEnt model and (F) the ensemble model. (G) Predicted potential distribution under RCP 8.5 for the MaxEnt model and (H) the ensemble model. (I) Change in probability of occurrence between the current and future potential distribution under RCP 8.5 for the MaxEnt model and (J) the ensemble model. Potential dis- tributions are given as the probability of occurrence. represent the environmental limitations of Hemidactylus species and environmental predictor in our models. A world digital elevation 2) their ability to conserve the climatic niche of a species. These vari- model was downloaded from www.ngdc.noaa.gov/mgg/topo/globe. ables included: annual mean temperature, maximum temperature of html. These data had a spatial resolution of 30 arc second (1 km) the warmest month, minimum temperature of the coldest month, and were resampled to the same resolution as the bioclim data using annual precipitation, precipitation of driest month, and precipita- bilinear interpolation. tion of the warmest quarter (Ro ¨ dder et al. 2008; Liu et al. 2017). In order to predict changes in potential species distribution Data with a spatial resolution of 5 arc minutes (10 km) was used. under a changing climate, we used bioclim variables predicted for In addition to climatic data, we also used elevation data as an the year 2050. These data were based on 11 CMIP5 climate Downloaded from https://academic.oup.com/cz/advance-article-abstract/doi/10.1093/cz/zox052/4101659 by Ed 'DeepDyve' Gillespie user on 08 June 2018 Weterings and Vetter Distribution of invasive house geckos 7 Figure 4. Results from the MaxEnt and ensemble models for H. frenatus based on current climate conditions and projected for 2050 using two representative con- centrations pathways (RCP). (A) Current potential distribution for the MaxEnt and (B) the ensemble model. (C) Predicted potential distribution under RCP 2.6 for the MaxEnt model and (D) the ensemble model. (E) Change in probability of occurrence between the current and future potential distribution under RCP 2.6 for the MaxEnt model and (F) the ensemble model. (G) Predicted potential distribution under RCP 8.5 for the MaxEnt model and (H) the ensemble model. (I) Change in probability of occurrence between the current and future potential distribution under RCP 8.5 for the MaxEnt model and (J) the ensemble model. Potential dis- tributions are given as the probability of occurrence. models (BCC-CSM1-1, CCSM4, GISS-E2-R, HadGEM2-AO, MaxEnt and ensemble models based on all climate models for each HadGEM2-ES, IPSL-CM5A-LR, MIROC-ESM-CHEM, MIROC- RCP. ESM, MIROC5, MRI-CGCM3, and NorESM1-M) for four repre- sentative concentration pathways (RCP 2.6, RCP 4.5, RCP 6, and Results RCP 8.5). Data from these models were resampled following Hijmans et al. (2005). We calculated the potential future distribu- Historical records showed that our studied species are widely dis- tion for each species based on all four RCPs for all 11 climate mod- tributed with some species overlapping in certain areas while other els. We then calculated the mean probability of occurrence for the areas are mainly occupied by single species (Figure 1). Hemidactylus Downloaded from https://academic.oup.com/cz/advance-article-abstract/doi/10.1093/cz/zox052/4101659 by Ed 'DeepDyve' Gillespie user on 08 June 2018 8 Current Zoology, 2017, Vol. 00, No. 00 Figure 5. Results from the MaxEnt and ensemble models for H. garnotii based on current climate conditions and projected for 2,050 using two representative con- centrations pathways (RCP). (A) Current potential distribution for the MaxEnt and (B) the ensemble model. (C) Predicted potential distribution under RCP 2.6 for the MaxEnt model and (D) the ensemble model. (E) Change in probability of occurrence between the current and future potential distribution under RCP 2.6 for the MaxEnt model and (F) the ensemble model. (G) Predicted potential distribution under RCP 8.5 for the MaxEnt model and (H) the ensemble model. (I) Change in probability of occurrence between the current and future potential distribution under RCP 8.5 for the MaxEnt model and (J) the ensemble model. Potential dis- tributions are given as the probability of occurrence. brookii is mostly found in south Asia, central Africa and northern neotropics and H. turcicus is currently found in most of the parts of South America. Hemidactylus frenatus is mainly found in Mediterranean region, Mexico, and the southern states of the South East Asia, northern Australia, Madagascar, Central America, United States. Most species showed little overlap in their actual dis- and many Pacific islands. Hemidactylus garnotii is mostly found in tribution with the other species, with H. garnotii as an exception. Southeast Asia, Florida, and various Pacific islands, Hemidactylus In total, 74.9% of all occurrence records did not co-occur with any mabouia in central and southern Africa and in most parts of the other species within a 0.5 degree (55.5 km) sampling grid Downloaded from https://academic.oup.com/cz/advance-article-abstract/doi/10.1093/cz/zox052/4101659 by Ed 'DeepDyve' Gillespie user on 08 June 2018 Weterings and Vetter Distribution of invasive house geckos 9 Figure 6. Results from the MaxEnt and ensemble models for H. mabouia based on current climate conditions and projected for 2050 using two representative concentrations pathways (RCP). (A) Current potential distribution for the MaxEnt and (B) the ensemble model. (C) Predicted potential distribution under RCP 2.6 for the MaxEnt model and (D) the ensemble model. (E) Change in probability of occurrence between the current and future potential distribution under RCP 2.6 for the MaxEnt model and (F) the ensemble model. (G) Predicted potential distribution under RCP 8.5 for the MaxEnt model and (H) the ensemble model. (I) Change in probability of occurrence between the current and future potential distribution under RCP 8.5 for the MaxEnt model and (J) the ensemble model. Potential distributions are given as the probability of occurrence. (Table 1). Hemidactylus turcicus showed the least overlap in its Africa and the Caribbean. Florida and Myanmar had the highest known historical distribution with other species (24.2%), while number of species being recorded within single grid cells. In H. garnotii was often recorded in grid cells where other species Myanmar, these were three native species (H. brookii, H. frenatus, where recorded as well (88.6%). Hemidactylus frenatus, and H. garnotii), while in Florida these were four invasive species H. mabouia, and H. brookii co-occurred with other species in re- (H. frenatus, H. garnotii, H. mabouia, and H. turcicus). spectively, 33.7%, 38.3%, and 52.9% of the cases. The largest over- Our models that predicted the potential distribution performed lap was between H. brookii and H. mabouia, where the southern well, with both the MaxEnt and ensemble models having AUC val- and northern edge of their historical distribution meet in Central ues higher than 0.88 (Table 2). The ensemble model for H. brookii Downloaded from https://academic.oup.com/cz/advance-article-abstract/doi/10.1093/cz/zox052/4101659 by Ed 'DeepDyve' Gillespie user on 08 June 2018 10 Current Zoology, 2017, Vol. 00, No. 00 Figure 7. Results from the MaxEnt and ensemble models for H. turcicus based on current climate conditions and projected for 2050 using two representative con- centrations pathways (RCP). (A) Current potential distribution for the MaxEnt and (B) the ensemble model. (C) Predicted potential distribution under RCP 2.6 for the MaxEnt model and (D) the ensemble model. (E) Change in probability of occurrence between the current and future potential distribution under RCP 2.6 for the MaxEnt model and (F) the ensemble model. (G) Predicted potential distribution under RCP 8.5 for the MaxEnt model and (H) the ensemble model. (I) Change in probability of occurrence between the current and future potential distribution under RCP 8.5 for the MaxEnt model and (J) the ensemble model. Potential dis- tributions are given as the probability of occurrence. had a lower kappa statistic (0.615) than all other ensemble models ensemble models, the general extend of the predicted potential dis- but this was still within the acceptable range. In general, ensemble tribution and predicted changes as an effect of climate change models resulted in larger areas that are considered suitable in com- where similar for both model types except for H. frenatus models parison to MaxEnt models (Table 3). Probabilities sampled at the lo- (Figures 3–7, Table 3). MaxEnt models predicted an increase in cation of the occurrence records also showed significant higher H. frenatus’ potential distribution, while the ensemble models pre- probabilities given by ensemble models in comparison to MaxEnt dicted a very small decrease. Nevertheless, the increase predicted by models (Figure 2). Although probabilities were generally higher for the MaxEnt model was mostly within regions also considered Downloaded from https://academic.oup.com/cz/advance-article-abstract/doi/10.1093/cz/zox052/4101659 by Ed 'DeepDyve' Gillespie user on 08 June 2018 Weterings and Vetter Distribution of invasive house geckos 11 Figure 8. Comparison of probability of occurrence from MaxEnt models for each Hemidactylus species on locations of occurrence records for (A) H. brookii,(B) H. frenatus,(C) H. garnotii,(D) H. mabouia, and (E) H. turcicus. P-values are given for each species when compared to the focal species (red box). If differences are non-signiﬁcant a given species’ probability of occurrence is similar on locations where the focal species (red box) occurred. suitable by the ensemble model. The minimum temperature of the temperature. Both models predicted a strong increase in suitable coldest month proved to be the most important variable for all spe- area under all four climate change scenarios. The area increase cies except for H. turcicus (Table 2). Hemidactylus turcicus’ distri- ranged from 46% to 58% under RCP 2.6 and 85% to 95% under bution was better explained by annual temperature. Annual RCP 8.5 in large parts of South and Central America as well as precipitation was more important in the H. garnotii model in com- Central Africa. The historical distribution of H. brookii was mostly parison with other species. Elevation was also more important for concentrated in Sub-Saharan Africa, India, Myanmar, Colombia, H. garnotii. and various islands of the Carribean. The potential distribution, For H. brookii the total area, with a probability of occurrence however, also covered large parts of tropical South and Central higher than 0.5, was 1.6 times higher for the MaxEnt model in com- America. parison to the ensemble model. This species had the largest potential Hemidactylus frenatus had the largest potential distribution distribution following the MaxEnt model but not for the ensemble (probability of occurrence> 0.50) for the ensemble model, which model. In both models, the minimum temperature of the coldest was 3.5 times higher than for the MaxEnt model. In both the month was the most important predictor followed by the mean MaxEnt and ensemble models, the minimum temperature of the Downloaded from https://academic.oup.com/cz/advance-article-abstract/doi/10.1093/cz/zox052/4101659 by Ed 'DeepDyve' Gillespie user on 08 June 2018 12 Current Zoology, 2017, Vol. 00, No. 00 Figure 9. Comparison of probability of occurrence from ensemble models for each Hemidactylus species on locations of occurrence records for (A) H. brookii,(B) H. frenatus,(C) H. garnotii,(D) H. mabouia, and (E) H. turcicus. P-values are given for each species when compared to the focal species (red box). If differences are non-signiﬁcant a given species’ probability of occurrence is similar on locations where the focal species (red box) occurred. coldest month was the most important variable. The mean tempera- Hemidactylus garnotii showed the smallest potential distribution ture was also important in the ensemble model but not in for both models. In both models, elevation was an important vari- the MaxEnt model. Precipitation of the hottest quarter was import- able as well as the minimum temperature of the coldest month. In ant in both models, although higher in the ensemble model. the MaxEnt model, precipitation played a major role as well (annual The historical distribution showed that H. frenatus was mostly con- precipitation and precipitation of hottest quarter), while in the en- centrated in South East Asia, Northern Australia, Madagascar, semble model the mean temperature was important. The historical Central America, Pacific islands, and Columbia. The potential distri- distribution showed that H. garnotii is mostly restricted to South bution also covered parts of Central Africa, East Coast of Africa, East Asia, Florida, and Pacific islands. The potential distribution did and the Amazon. Under a changing climate, the probability of oc- not deviate much from the historical distribution except for the currence was predicted to decline in much of its current invasive Pampas (lowland Argentina and Uruguay). Both models predicted a range, but will expand outwards from the equator into new territo- decline ranging between 17% to 30% for RCP 2.6 and 27% to 36% ries such as South Australia and the Gulf Coast of the United States. for RCP 8.5. Under these climate-change scenarios, H. garnotii’s Downloaded from https://academic.oup.com/cz/advance-article-abstract/doi/10.1093/cz/zox052/4101659 by Ed 'DeepDyve' Gillespie user on 08 June 2018 Weterings and Vetter Distribution of invasive house geckos 13 Table 4. Area covered by individual species and species groups relative to the total global area for which the probability was higher than 50% for at least one species. Species groups which covered less than 1% of the total area are excluded from the table. Results are given for all ﬁve focal hemidactylus species. Species (groups) Current RCP 2.6 RCP 8.5 MaxEnt (%) Ensemble (%) MaxEnt (%) Ensemble (%) MaxEnt (%) Ensemble (%) H. brookii 35.3 9.4 46.7 13.6 53.6 17.5 H. turcicus 20.2 39.2 17.2 29.6 16.8 29.7 H. frenatus 11.5 31.8 11.5 26.0 6.5 25.5 H. brookii and H. frenatus 4.7 5.0 9.3 9.6 11.7 12.1 H. frenatus and H. mabouia 1.3 24.8 0.4 7.5 0.1 4.2 H. mabouia 8.7 8.9 3.4 3.0 2.1 2.0 H. brookii, H. frenatus and H. mabouia 4.7 8.9 2.8 4.2 1.3 3.2 H. brookii and H. mabouia 7.5 3.9 4.3 2.5 3.1 2.1 native range was predicted to become less suitable, but parts of the were all other species were recorded. For H. turcicus, the conditions Gulf Coast of the United States, the Pampas, and New Zealand’s were also unsuitable where other species were sampled but were North Island are predicted to become more suitable. slightly better on H. garnotii sites. The MaxEnt models showed that The ensemble model for H. mabouia showed a 2.4 times larger from all species the potential area for H. brookii covered the largest potential area (probability of occurrence> 0.5) than the MaxEnt area that was not suitable (probability< 0.5) for any other species fol- model. The minimum temperature of the coldest month was most lowed by H. turcicus (Table 4). For the ensemble models, the largest important in the MaxEnt model, for the ensemble model this was area that was only suitable for one species was for H. turcicus followed the mean temperature followed by the minimum temperature of the by H. frenatus. Both models also showed that the potential distribution coldest month. The historical distribution showed that the species is of H. brookii and H. frenatus covered large areas that were suitable for mostly concentrated in East and West Africa, tropical South both species. The ensemble models also showed that large areas were America, and the Caribbean. The potential distribution also shows suitable for both H. frenatus and H. mabouia. suitable areas in South East Asia and East Coast Australia. Under a changing climate the potential distribution was predicted to reduce with 41% to 44% under RCP 2.6 and by 55% to 60% under RCP Discussion 8.5. This decrease will cover most of its current potential range. The ensemble model for H. turcicus also returned a larger suit- Our study showed that most of the focal species have already able area than the MaxEnt model. The ensemble model resulted in a invaded large parts of their potential distribution, but do not cover 2.3 times larger area (probability of occurrence> 0.5) than the all areas that are potentially suitable. Large areas suitable for Maxent model. In both models, the mean temperature was the most H. brookii and H. frenatus in South America are not colonized by important variable followed by the minimum temperature of the these species yet. In Continental Africa, H. frenatus is mostly absent, coldest month for the MaxEnt model and annual precipitation and while climatic conditions are favourable. Most of these areas that minimum temperature of the coldest month respectively for the en- are highly suitable for one or more species but have not yet been semble model. Along the coastal areas of its current potential distri- colonized by these species often contain other well established spe- bution, climatic conditions appear to become slightly more suitable. cies. Areas where H. brookii or H. frenatus is absent are often colon- The historical distribution showed that the species is mostly found ized by H. mabouia and vice versa. For example, in Suriname (South not only in arid and generally colder regions such as its native range America) H. mabouia is the most commonly distributed species surrounding the Mediterranean Sea and in its invasive range in the (Nielsen et al. 2013), even though this area is also highly suitable for American Mid-West and California but also in Florida. The models H. brookii and H. frenatus. In Ghana on the other hand, areas that also showed that its potential range extends to other areas such as are highly suitable for H. mabouia and H. frenatus are dominated the arid regions of Argentina, South Africa, and South Australia. by H. brookii (Leache ´ 2005). This segregated distribution of these Both models showed that the area of its potential distribution will species may have several causes. First, certain species have not yet increase from 8% to 10% under RCP 2.6 and from 17% to 27% been introduced to these areas. However, this is unlikely considering under RCP 8.5. The distribution will mostly increase along the the wide scale of anthropogenic introduction of these species over northern border of its current distribution within the northern hemi- the last few centuries. Second, certain species are better competitors sphere and along the southern border within in the southern hemi- under certain environmental conditions than other species (Ro ¨ dder sphere, while the suitability decreases in large parts of central et al. 2008). When environmental conditions are more favorable for Australia, the African Mediterranean coast, and along the Gulf a given species their fitness increases, giving the species a competi- Coast of the United States. tive advantage over the other species. While under slightly different Conditions for H. brookii on locations where H. frenatus and environmental conditions the competitive advantage may shift H. mabouia were recorded were generally slightly less suitable, but for towards the other species. For example, in the native range of H. garnotii and H. turcicus locations the conditions were unsuitable H. brookii and H. frenatus the tokay gecko (Gekko gecko)isan im- (Figures 8 and 9). Conditions for H. frenatus and H. mabouia were portant predator of house geckos (personal observations). equally suitable at locations where H. garnotii was recorded. Behavioral differences may affect the rate on which house geckos Conditions for H. garnotii were highly unsuitable on most locations are preyed upon and this may strongly affect community structure Downloaded from https://academic.oup.com/cz/advance-article-abstract/doi/10.1093/cz/zox052/4101659 by Ed 'DeepDyve' Gillespie user on 08 June 2018 14 Current Zoology, 2017, Vol. 00, No. 00 (Abrams and Matsuda 1993; Schmitz 2005). In the absence of this Similar to other studies in which both MaxEnt and ensemble mod- predator other species specific traits may favor H. brookii instead. els were used, our ensemble models generally tended to have higher probabilities of occurrence on the locations of the historical occurrence As an effect of climate change, regions that are suitable for records in comparison to the MaxEnt models (Simpson et al. 2011; H. brookii, H. frenatus, and H. turcicus will increase. The SDMs Baier et al. 2014; Ihlow et al. 2016; Ashraf et al. 2017). This often re- showed that H. brookii and H. turcicus have the highest tolerance to sults in larger areas that are suitable for a given species (Simpson et al. extreme temperatures. Hemidactylus turcicus can tolerate colder 2011; Ihlow et al. 2016; Ashraf et al. 2017). Ensemble models tend to temperatures than the other Hemidactylus species (Huey et al. be more accurate for modeling the current distribution but perform 1989). Hemidactylus brookii seems to tolerate higher temperatures less outside the range where occurrence data are sampled (Ashraf et al. than the other species. This gives these two species the advantage 2017). While on the other hand, MaxEnt models have a higher predict- that allows them to expand their range into regions that are less ive performance (Elith and Leathwick 2009). suitable to other species. An increase in suitable area does not neces- In conclusion, as an effect of global climate change it is likely that sarily mean that these species will occupy these areas in the near fu- H. brookii will expand its range to areas that are currently colonized ture. Hemidactylus species do not easily disperse over large by H. mabouia and/or H. frenatus. As these areas are already invaded distances by themselves, therefore, dispersal is very much restricted by other Hemidactylus species it is likely that the expansion of to human mediation. In Florida, H. mabouia has been shown to be H. brookii will only have minimal ecological impact on these areas. genetically homogeneous across the state, indicating that human- Of greater concern is the species H. turcicus, which is likely to expand mediated dispersal is frequent (Short and Petren 2011). However, its range to areas that are not yet invaded by any Hemidactylus species. on a global scale, Short and Petren (2011) showed that human- More research is needed to elucidate the effects of climate change on mediated dispersal is less common. Hemidactylus species are often interspecies interactions. Especially interactions between H. brookii found in and on cars by which they easily disperse long distances by and H. frenatus are important because the overlap in suitable area of road (Norval et al. 2012). Hence, regions in which humans move these two species will double. The effects of interspecies interactions easily over large distances by road are therefore likely to become may severely affect the distribution of Hemidactylus species. more quickly invaded by Hemidactylus species than regions that Knowledge about these mechanisms may also provide further insight are more isolated. For example, H. brookii is now widely spread in the impacts of climate change on ecological systems in context of in Sub-Saharan Africa and is likely to extend its range further south species distributions and territory expansion. into areas currently dominated by H. mabouia for which the envir- onmental suitability will strongly decrease. However, in South America H. brookii is less widely established and the de- crease in suitable area for H. mabouia does not necessarily mean Funding H. brookii can easily occupy these regions. Currently, H. brookii is This study was funded by the Cat Drop Foundation. only present north of the Amazons, where the Amazons form a barrier and hinder easy dispersal mediated by anthropogenic activ- ities. However, vast deforestation of the Amazons will increase trans-Amazon human movement and eventually this will mediate Acknowledgments the dispersal of H. brookii into larger parts of South America We would like to thank the anonymous reviewers for their constructive com- as well. ments that have helped to substantially improve the article. Under a changing climate, H. turcicus is likely to expand its dis- tribution vastly throughout North America and Europe. In these re- References gions, human mobility is high (Hawelka et al 2014) and a lack of natural barriers will easily facilitate the dispersal of this species. In Abrams P, Matsuda H, 1993. Effects of adaptive predatory and anti-predator Australia, where the species is absent but the environment is highly behaviour in a two-prey-one-predator system. Evol Ecol 7:312–326. suitable, the species is likely to remain absent due to very strict bio- Allouche O, Tsoar A, Kadmon R, 2006. Assessing the accuracy of species dis- security regulations (Nelson et al. 2014). However, if the species tribution models: prevalence, kappa and the true skill statistic (TSS). J Appl Ecol 42:1223–1232. does establish in Australia it will likely become more widespread Ashraf U, Peterson A, Chaudhry MN, Ashraf I, Saqib Z et al. 2017. Ecological than the already present invasive H. frenatus. niche model comparison under different climate scenarios: a case study of The distribution of H. garnotii shows a tendency to decline in Olea spp. in Asia. Ecosphere 8:e01825. much of its native range. However, along the Gulf Coast of the Baier F, Nicolaou H, Ro ¨ dder D, 2014. A species distribution model for the en- United States, the climate will become more suitable, here the spe- demic Cyprus whip snake Hierophis cypriensis is consistent with a transient cies is likely to replace H. turcicus. H. garnotii, is stronger in fights period of isolated evolution in the Troodos Range. Herpetol J 24:175–181. and currently already replaces H. turcicus in Florida due to this Bauer AM, Jackman TR, Greenbaum E, Giri VB, de Silva A, 2010. South Asia competitive advantage (Frankenberg 1984; Meshaka Jr 1995). supports a major endemic radiation of Hemidactylus geckos. Mol However, this same region will also become more suitable for Phylogenet Evol 57:343–352. Butchart SHM, Walpole M, Collen B, van Strien A, Scharlemann JPW et al. H. frenatus. This species is already well established in the south of 2010. Global biodiversity: indicators of recent declines. Science 328: Florida and is likely to spread along the coast facilitated by the high 1164–1168. mobility of people. In areas where H. garnotii and H. frenatus co- Callaway Z, Thomas A, Melrose W, Buttner P, Speare R, 2011. Salmonella occur, the smaller H. frenatus manages to suppress H. garnotii Virchow and Salmonella Weltevreden in a random survey of the Asian through sexual interference (Dame and Petren 2006). Hence, it is house gecko Hemidactylus frenatus in houses in Northern Australia. likely that in due time, populations of H. garnotii will also become Vector-Borne Zoon Diseases 11:621–625. smaller in this region. In addition our models showed that Carranza S, Arnold EN, 2006. Systematics, biogeography, and evolution of Hemidactylus garnotii is mostly restricted to lowland areas, which Hemidactylus geckos (Reptilia: Gekkonidae) elucidated using mitochon- pose a barrier to further increase its distribution (Zug et al. 2007). drial DNA sequences. Mol Phylogenet Evol 38:531–545. Downloaded from https://academic.oup.com/cz/advance-article-abstract/doi/10.1093/cz/zox052/4101659 by Ed 'DeepDyve' Gillespie user on 08 June 2018 Weterings and Vetter Distribution of invasive house geckos 15 Cook DC, Thomas MB, Cunningham SA, Anderson DL, De Barro PJ, 2006. Meshaka WE, Jr, 2000. Colonization dynamics of two exotic geckos Predicting the economic impact of an invasive species on an ecosystem ser- (Hemidactylus garnotii and H. mabouia) in Everglades National Park. vice. Ecol Appl 17:1832–1840. J Herpetol 34:163–168. Dame EA, Petren K, 2006. Behavioural mechanisms of invasion and displace- Microsoft, 2017. Bing Maps. Tampa, Florida: Microsoft. ment in Paciﬁc island geckos Hemidactylus. Animal Behav 71:1165–1173. Murphy-Walker S, Haley SR, 1996. Functional sperm storage duration in fe- Elith J, Leathwick JR, 2009. Species distribution models: ecological explan- male Hemidactylus frenatus (family Gekkonidae). Herpetologica 52: ation and prediction across space and time. Ann Rev Ecol Evol Syst 40: 365–373. 677–697. Nelson M, Roffey P, McNevin D, Lennard C, Gahan ME, 2014. An overview Elith J, Phillips SJ, Hastie T, Dud´k ı M, Chee YE et al. 2011. A statistical ex- of biosecurity in Australia. Aus J Forensic Sci 46:383–396. planation of MaxEnt for ecologists. Divers Distrib 17:43–57. Newbery B, Jones D, 2007. Presence of Asian house gecko Hemidactylus fre- ESRI, 2012. ArcGIS Desktop: Release 10.1. Environmental Systems Research natus across an urban gradient in Brisbane: inﬂuence of habitat and poten- Institute, Redlands, California. tial for impact on native gecko species. In: Lunney D, Eby P, Hutchings P, Fourcade Y, Engler JO, Rodder D, Secondi J, 2014. Mapping species distribu- Burgin S, editors. Pest or Guest: the Zoology of Overabundance. Mosman, tions with MAXENT using a geographically biased sample of presence data: Australia: Royal Zoological Society of New South Wales. 59–65. a performance assessment of methods for correcting sampling bias. PLoS Nielsen S, Jairam R, Ouboter P, Noonan B, 2013. A herpetofaunal survey ONE 9:1–13. of the Grensgebergte and Kasikasima regoins, Suriname. In: A Rapid Frankenberg E, 1984. Interactions between two species of colonizing house Biological Assessment of the Upper Palumeu River Watershed geckos Hemidactylus turcicus and Hemidactylus garnotii. J Herpetol 18: (Grensgebergte and Kasikasima) of Southeastern Suriname. 131–144. 1–7. Norval G, Lu FY, Mao JJ, Slater K, 2012. It is not inside, it is on top! An ex- Gurevitch J, Padilla DK, 2004. Are invasive species a major cause of extinc- ample of vehicular-rafting by a house gecko (Hemidactylus frenatus tions?. Trends Ecol Evol 19:470–474. Schlegel, 1836). Herpetol Notes 5:451–452. Hawelka B, Sitko I, Beinat E, Sobolevsky S, Kazakopoulos P et al. 2014. Petren K, Case TJ, 1996. An experimental demonstration of exploitation com- Geo-located Twitter as proxy for global mobility patterns. Cartogr Geogr petition in an ongoing invasion. Ecology 77:118–132. Inform Sci 41:260–271. Phillips SJ, Anderson RP, Schapire RE, 2006. Maximum entropy modeling of Heinsohn T, 2003. Animal translocation: long-term human inﬂuences on the species geographic distributions. Ecol Model 190:231–259. vertebrate zoogeography of Australasia (natural dispersal versus ethnophor- Pimentel D, Zuniga R, Morrison D, 2005. Update on the environmental and esy). Aus Zool 32:351–376. economic costs associated with alien-invasive species in the United States. Hellmann JJ, Byers JE, Bierwagen BG, Dukes JS, 2008. Five potential Ecol Econ 52:273–288. consequences of climate change for invasive species. Conserv Biol 22: Ro ¨ dder D, Lo ¨ tters S, 2009. Niche shift versus niche conservatism? Climatic 534–543. characteristics of the native and invasive ranges of the Mediterranean house Hoskin CJ, 2011. The invasion and potential impact of the Asian house gecko gecko (Hemidactylus turcicus). Global Ecol Biogeogr 18:674–687. Hemidactylus frenatus in Australia. Aus Ecol 36:240–251. Ro ¨ dder D, Sole´M, Bo ¨ hme W, 2008. Predicting the potential distributions of Huey RB, Niewiarowski PH, Kaufmann J, Herron JC, 1989. Thermal biology two alien invasive house geckos (Gekkonidae: Hemidactylus frenatus, of nocturnal ectotherms: is sprint performance of geckos maximal at low Hemidactylus mabouia). North-Western J Zool 4:236–246. body temperatures. Physiol Zool 62:488–504. R Development Core Team, 2017. R: A Language and Environment for Statistical Ihlow F, Courant J, Secondi J, Herrel A, Rebelo R et al. 2016. Impacts of cli- Computing. Vienna, Austria: R Foundation for Statistical Computing. mate change on the global invasion potential of the African clawed frog RStudio, 2013. RStudio: Integrated Development Environment for R (Version Xenopus laevis. PLoS ONE 11:e0154869. 0.98.1103). Boston, MA: RStudio. Iturriaga M, Marrero R, 2013. Feeding ecology of the tropical house gecko Selcer KW, 1986. Life history of a successful colonizer: the Mediterranean Hemidactylus mabouia (Sauria : Gekkonidae) during the dry season in Gecko Hemidactylus turcicus, in Southern Texas. Copeia 4:956–962. Havana, Cuba. Herpetol Notes 6:11–17. Schmitz OJ, 2005. Behavior of predators and prey links with population-level Jeschke JM, Strayer DL, 2008. Usefulness of bioclimatic models for studying processes. In: Barbosa P, Castellanos I, editors. Ecology of Predator Prey climate change and invasive species. Ann NY Acad Sci 1134:1–24. Interactions. New York: Oxford University Press. 256–278. Krysko KL, Sheehy IIICM, Hooper AN, 2003. Interspeciﬁc communal ovipos- Short KH, Petren K, 2011. Multimodal dispersal during the range expansion ition and reproduction of four species of lizards (Sauria: Gekkonidae) in the of the tropical house gecko Hemidactylus mabouia. Ecol Evol 1:181–190. Lower Florida Keys. Amphibia-Reptilia 24:390–396. Simpson M, Prots B, Vykhor B, 2011. Modeling of the invasive plant distribu- Kurita T, 2013. Current status of the introduced common house gecko tion: case study of Sosnowski’s hogweed Heracleum sosnowski Manden in Hemidactylus frenatus (Squamata: Gekkonidae) on Amamioshima Island of the Ukranian Carpathian Mts. J Biol Syst 3:80–89. the Ryukyu Archipelago, Japan. Curr Herpetol 32:50–60. Swets JA, 1988. Measuring the accuracy of diagnostic systems. Science 240: Leache ´ A, 2005. Results of a herpetological survey in Ghana and a new coun- 1285–1293. try record. Herpetol Rev 36:16–19. Thuiller W, Lafourcade B, Engler R, Araujo MB, 2009. BIOMOD: a platform Li X, Liu X, Kraus F, Tingley R, Li Y, 2016. Risk of biological invasions is for ensemble forecasting of species distribution. Ecography 32:369–373. concentrated in biodiversity hotspots. Front Ecol Env 14:411–417. Tkaczenko GK, Fischer AC, Weterings R, 2014. Prey preference of the com- Li Y, Liu X, Li X, Petitpierre B, Guisan A, 2014. Residence time, expansion to- mon house geckos Hemidactylus frenatus and Hemidactylus platyurus. ward the equator in the invaded range and native range size matter to cli- Herpetol Notes 7:483–488. matic niche shifts in non-native species. Global Ecol Biogeogr 23: Weterings R, 2017. Observations of an opportunistic feeding strategy in 1094–1104. ﬂat-tailed house geckos Hemidactylus platyurus living in buildings. Liu X, Petitpierre B, Broennimann O, Li X, Guisan A et al. 2017. Realized cli- Herpetol Notes 10:133–135. matic niches are conserved along maximum temperatures among herpeto- Wickman H, 2009. ggplot2: Elegant Graphics for Data Analysis (Use R!). 1st faunal invaders. J Biogeogr 44:111–121. edition. Dordrecht, NL: Springer. Meshaka WE, Jr, 1995. Reproductive cycle and colonization ability of the Zug GR, Vindum JV, Koo MS, 2007. Burmese Hemidactylus (Reptilia, Mediterranean gecko Hemidactylus turcicus in south-central Florida. Squamata, Gekkonidae): taxonomic notes on tropical Asian Hemidactylus. Florida Sci 58:10–15. Proceedings of the California Academy of Sciences. 58: 387–405. Downloaded from https://academic.oup.com/cz/advance-article-abstract/doi/10.1093/cz/zox052/4101659 by Ed 'DeepDyve' Gillespie user on 08 June 2018
Current Zoology – Oxford University Press
Published: Sep 1, 2017
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera