Genetic Algorithm For Rule-set Prediction Research Articles

Changes in Overwintering Ground of Small Yellow Croaker (Larimichthys polyactis) Based on MaxEnt and GARP Models: A Case Study of the Southern Yellow Sea Stock.

Small yellow croaker (Larimichthys polyactis), a commercially essential fish commonly caught in China and South Korea, is now facing a severe decline in resources. The recruitment and surplus of L. polyactis depend selecting a suitable marine environment for overwintering. However, the international overwintering migration habit of L. polyactis limits the investigation of its overwintering environment preferences and suitable grounds. In this study, based on the distribution data of L. polyactis in the southern Yellow Sea in winter from 2010 to 2019 and the ocean remote sensing data such as sea bottom temperature (SBT), sea bottom salinity, chlorophyll-a concentration and water depth (Depth), we used the maximum entropy (MaxEnt) and the genetic algorithm for rule-set production (GARP) models to investigate the overwintering grounds of the southern Yellow Sea stock (SYS). The jackknife test was used to assess the importance of various environmental factors. For modelling the overwintering ground distribution of SYS, the area under the curve values of both models ​​were higher than 0.9. The overwintering ground was at 32°10'N-33°48'N, 122°30'E-125°00'E. The direction of its distribution was consistent with the Yellow Sea Warm Current in the southern Yellow Sea during the winter. Compared with the suitable overwintering area during 2010-2014, the highly appropriate overwintering area for SYS to overwinter decreased significantly during 2015-2019, showing a trend of moving to the east and north, related to the increase in fishing pressure and strengthening of the Yellow Sea Warm Current in recent years. Depth was the most significant factor for SYS overwintering, followed by SBT. The overwintering ground was at a depth of 40-65 m during the two periods. Additionally, the suitability of overwintering grounds in the coastal waters of southwestern South Korea has gradually increased. This study provides a scientific basis for formulating effective strategies to manage L. polyactis resources under the China-South Korea Fisheries Agreement. This article is protected by copyright. All rights reserved.

Coming home: modelling the mating roost of the endangered bat Leptonycteris nivalis

The Mexican Long-nosed bat (Leptonycteris nivalis) is the largest nectarivorous species in the New World, and one of three migratory nectarivores in Mexico. It is considered an ‘Endangered Species’ under the U.S. Endangered Species Act and ‘Threatened’ by the Mexican Federal List of Endangered Species. In 1994, a Recovery Plan was developed by the USFWS with the participation of Mexican and American researchers, and the most urgent actions to ensure the species protection were identified. Locating and protecting roosts are among the most urgent tasks recognized. With this study, we aimed to identify the most suitable areas potentially holding additional mating roosts of Leptonycteris nivalis, and we conducted surveys of these areas to confirm its presence, and to assess the reproductive state of individuals. We used Maxent, the Genetic Algorithm for Rule-set Production (GARP), and Bioclim algorithms to generate an agreement map of the potential distribution of additional mating roosts, and we implemented a Euclidian multidimensional distances analysis to identify ecologically similar regions to “La Cueva del Diablo”, the only mating roost known for the species. We identified suitable areas in the states of Morelos, Puebla and the State of Mexico. We visited seventeen caves distributed in ten different localities in these areas. For two consecutive years, we found the species in a cave called: “La Cueva de los Coyotes”, located in the State of Mexico, where we captured eighteen individuals, including a pregnant female. The location of an unknown roost so far, occupied by individuals of L. nivalis, and among them a pregnant female, allows us to reflect about the reproductive dynamics of the species. In that sense, reproductive populations may be splitting into smaller colonies to mate, other than “La Cueva del Diablo”, or pregnant females might me moving to additional and nearby roosts to spend the rest of the winter season. Using these tools and further refinements we may be able to locate additional mating roosts, thus, providing more possibilities for the application of conservation measures for the protection of the species.

Моделирование распространения, численности и выживаемости видов: новые возможности и методы

Моделирование распространения, численности и выживаемости видов: новые возможности и методы

Modeling the Susceptibility of Forest Fires Using a Genetic Algorithm: A Case Study in Mountain Areas of Southwestern China

Modeling fire susceptibility in fire-prone areas of forest ecosystems was essential for providing guidance to implement prevention and control measures of forest fires. Traditional models were developed on the basis of random selection of absence data (i.e., nonfire data from unburned areas), which could bring uncertainties to modeling results. Here, a new model with the genetic algorithm for Rule-set Production (GARP) algorithm and 10 environmental layers was proposed to process presence-only data in the susceptibility modeling of forest fires in Chongqing city. To do this, 70% of 684 fire occurrence data (479) during the period of 2000–2018 were applied to train the proposed model. And, 30% of these fire occurrence data (205) and the same amount of no-fire data (205) were emerged as validation dataset. The results showed that, for some environmental layers (i.e., distance to the nearest road, land cover, precipitation, distance to the nearest settlement, aspect, relative humidity, elevation, wind speed, and temperature), their P values were less than 0.05, indicating that these 9 environmental layers have significant influence on the spatial distribution of fire susceptibility in Chongqing city. On the contrary, with a higher P value (i.e., 0.126), the slope layer has an insignificant effect on fire susceptibility in the study area. Furthermore, the results of receiver operating characteristic analysis (ROC) showed that the proposed model has a good performance with an AUC value of 0.869, an accuracy value of 0.732, a sensitivity value of 0.59, a specificity value of 0.873, a positive predictive value of 0.823, and a negative predictive value of 0.681. This study revealed the validity of the proposed model in modeling the susceptibility of forest fires.

GARPTools: R software for data preparation and model evaluation of GARP models

The aim of the GARPTools package is to provide tools to prepare data for input into the desktop version of the genetic algorithm for rule‐set prediction (GARP), for the evaluation of the accuracy of output models, and for summary/examination of environmental coverages used in GARP rule sets for best models in an experiment. GARP is a software package for biodiversity and ecological research that allows the user to predict and analyze wild species' geographic distributions. GARP is a presence‐background genetic algorithm that models species' potential geographic distributions through an iterative process of training and testing that occurs through resampling and replacement of input data. GARP develops rule‐sets of if/then logic statements that assign presence or absence, which are then mapped on the landscape. Toward this, GARPTools provides preparation functions including splitting species presence locations into training and testing data sets and resampling environmental layers to the same spatial resolution (raster cell size) and extent. Model evaluation functions are relevant to current procedures applied to species distribution modeling, including the receiver operating characteristic curves and omission and commission indices. There are also functions to estimate the contribution of environmental coverages (covariates) based on the GARP outputs by examining the individual rules. GARPTools intends to provide a means to systematically prepare data, evaluate models, and summarize environmental coverages among multiple systems and species without the need for complex software programming.

A deadly encounter: Alien invasive Spodoptera frugiperda in Africa and indigenous natural enemy, Cotesia icipe (Hymenoptera, Braconidae).

The invasion and wide spread of Spodoptera frugiperda represent real impediments to food security and the livelihood of the millions of maize and sorghum farming communities in the sub-Saharan and Sahel regions of Africa. Current management efforts for the pest are focused on the use of synthetic pesticides, which are often economically unviable and are extremely hazardous to the environment. The use of biological control offers a more economically and environmentally safer alternative. In this study, the performance of the recently described parasitoid, Cotesia icipe, against the pest was elucidated. We assessed the host stage acceptability by and suitability for C. icipe, as well as its ovigenic status. Furthermore, the habitat suitability for the parasitoid in the present and future climatic conditions was established using Maximum Entropy (MaxEnt) algorithm and the Genetic Algorithm for Rule‐set Prediction (GARP). Cotesia icipe differentially accepted the immature stages of the pest. The female acceptance of 1st and 2nd instar larvae for oviposition was significantly higher with more than 60% parasitism. No oviposition on the egg, 5th and 6th larval instars, and pupal stages was observed. Percentage of cocoons formed, and the number of emerged wasps also varied among the larval stages. At initial parasitism, parasitoid progenies, time to cocoon formation and overall developmental time were significantly affected by the larval stage. Egg-load varied significantly with wasp age, with six-day-old wasps having the highest number of mature eggs. Ovigeny index of C. icipe was 0.53. Based on the models, there is collinearity in the ecological niche of the parasitoid and the pest under current and future climate scenarios. Eastern, Central and parts of coastal areas of western Africa are highly suitable for the establishment of the parasitoid. The geographic distribution of the parasitoid would remain similar under future climatic conditions. In light of the findings of this study, we discuss the prospects for augmentative and classical biological control of S. frugiperda with C. icipe in Africa.

Diversity, Host Plants and Potential Distribution of Edible Saturniid Caterpillars in Kenya.

The promotion of edible insects, including saturniid caterpillars as potential food source is widely gaining momentum. They are adequately rich in nutrients such as proteins, amino acids, fatty acids, and micronutrients. Despite saturniids being a traditional food source with economic benefits, information on their diversity, host plants and their potential distribution in Africa are lacking, which this study seeks to address. Edible saturniids and their host plants were characterized using specific primers (LepF1/LepR1 and 3F_KIM_F/1R_KIM_R, respectively). Maximum entropy (MaxENT) and GARP (genetic algorithm for ruleset production) models were used to characterize the potential distribution of commonly consumed saturniids under current and future climate scenarios. Seven species of saturniids were recorded from 11 host plants in Kenya: Gonimbrasia zambesina, Gonimbrasia krucki, Bunaea alcinoe, Gonimbrasia cocaulti, Gonimbrasia belina, Gynanisa nigra and Cirina forda. Two morphotypes of G. zambesina and B. alcinoe were recorded. These saturniid caterpillars occur twice a year except for G. cocaulti. Predictive models revealed that tropical and subtropical regions were potentially suitable for B. alcinoe and C. forda. The information generated from this study would be important to guide conservation efforts and their sustainable utilization as food in Africa.

Eco-climatic matching to guide foreign exploration and optimal release strategies for biological control agents of Rastrococcus iceryoides in Africa and Asia

Rastrococcus iceryoides (Green) (Homoptera: Pseudococcidae) is a major invasive pest of several horticultural crops [in Africa and Asia, outside its native range in India], with damage levels ranging from 30% to complete crop failure. Due to lack of effective co-evolved parasitoids in the invaded regions, maximum entropy (MaxEnt) and genetic algorithm for ruleset production (GARP) were used to identify climatically suitable areas in India for foreign exploration. Based on the outcome of the predictive models, an extensive survey was conducted in 15 major mango growing regions in the state of Tamil Nadu, India. Thereafter, both models were used to identify climatic compatibility habitats in the invaded regions of R. iceryoides. Our results revealed ten host plants belonging to eight families with considerably low levels of infestation. The percentage parasitism established using mummified R. iceryoides was relatively high ranging between 16.7 ± 1.4 to 91.3 ± 3.7%. Both old and new host-parasitoid associations were recorded with eleven parasitoid species described. Eight of the parasitoids recorded were new records of R. iceryoides. Among these parasitoids, Praleurocerus viridis Agarwal, Anagyrus chryos Noyes & Hayat and Neoplatycerus tachikawai Subba Rao were the most dominant and widespread parasitoid species, highly specific to R. iceryoides with percent parasitism of 53.2 ± 5.4, 31.3 ± 2.7 and 8.8 ± 2.9%, respectively. Using the occurrence data of the parasitoids, both models successfully identified optimal suitable habitats in Africa and Asia. Both models showed optimal performances with the value of the average area under the curve (AUC) of 0.98 for MaxEnt and 0.95 for GARP. However, the percentage contribution of the predictor variables that influenced the current and future predictions in the native and invaded range varied considerably. These findings demonstrate the importance of predictive modelling as novel tools to support future classical biological control program targeting R. iceryoides in the invaded regions. Our results provide important information to guide strategic planning for future classical biological control programmes.

Mapping the potential of mangrove forest restoration based on species distribution models: A case study in China

Mangrove forests support numerous ecosystem services and contribute to coastal ecological risk reduction. However, they are one of the most severely threatened ecosystems in the world. China has carried out national mangrove restoration projects, but there is still insufficient scientific information for the strategic planning of this restoration. In this study, we carried out mangrove suitability assessments using the genetic algorithm for rule-set prediction (GARP) and maximum entropy (MaxEnt) models, and we mapped the restoration potential of mangrove forests in China for the first time. The restoration potential index (RPI), which combines suitability and land use data, is proposed as a rapid estimator method for locating theoretically available areas for restoration. The results showed that the MaxEnt model performed better than GARP in predicting potential mangrove distributions. Temperature was the most important environmental factor for determining large scale distribution of mangroves. The predicted northern limit of mangrove distribution was around 28°27′ N–28°35′ N. Using the RPI approach, 16,800 ha with the potential to restore mangrove forests was identified. According to both models, the largest area with restoration potential occurs along the Guangdong and Guangxi coast. Nationwide, about 75% of the potential area suitable for mangrove forests has been lost as a consequence of land use and is no longer available for restoration. Around 6400 ha of ponds is currently used for aquaculture, accounting for 38% of theoretically restorable areas. These areas can be a priority for mangrove forest restoration. In conclusion, our findings provide a better scientific understanding of mangrove distribution in China and can underpin strategic design and planning of mangrove restoration.

Effect of climate change on potential distribution of Cedrus libani A. Rich in the twenty-first century: an Ecological Niche Modeling assessment

The present study is focused on the potential distribution of the Lebanese cedar (Cedrus libani) in the present and in the future throughout the twenty-first century. The location of this work encompasses Lebanon, Syria and Turkey. Twenty-four environmental variables are used and two Representative Concentration Pathways (RCP) scenarios for two different time periods are studied: RCP 4.5 2050, RCP 4.5 2070, RCP 8.5 2050 and RCP 8.5 2070. The most interesting novelty is the use of 13 General Circulation Models and 6 algorithms (Climate Space Model, Envelope Score, Environmental Distance, Genetic Algorithm for Rule-set Production, Maximum Entropy and Support Vector Machines) were considered for modelling. Area Under the Curve is used as goodness of fit and building the final consensus map. The global habitat suitability area would enlarge in the forecasted scenarios with respect to the present, although it would be more restricted in 2070 due to the altitudinal shift. This study also suggests an interesting approach to manage C. libani stands by means of afforestation programs aiming to face global warming in the late twenty-first century.

Effect of spatial resolution, algorithm and variable set on the estimated distribution of a mammal of concern: the squirrel Sciurus aberti

ABSTRACT Most potential habitat models have been built from WorldClim using low resolution variables, even for areas of high heterogeneity with few weather stations. The resulting models can be too general and lead to erroneous decisions when used for conservation purposes. Sciurus aberti is a tree squirrel inhabiting highlands in the SW US and the Sierra Madre Occidental (SMO) in Mexico, where it is considered a species of low concern. We examined the effect of resolution, variables, and algorithms on the predicted potential habitat of S. aberti in Mexico and compared the resulting models against a previous one created from WorldClim variables using GARP (Genetic Algorithm for Rule Set Production). Our best model, using Maxent, 30 m spatial resolution and topographic variables, predicted a fragmented distribution in pine and pine–oak forests, consistent with what is known about the species’ natural history. The area represented only 2% of the SMO (compared to 28% for the GARP model), of which only 0.33% lies within protected areas. The model suggests that the habitat is highly fragmented, which threatens population continuity. Therefore, we propose that the conservation status of Sciurus aberti must be reassessed and that forest management better consider the conservation of arboreal species.

Species Distribution Modeling of Sassafras Tzumu and Implications for Forest Management

Sassafras tzumu (Chinese sassafras) is an economically and ecologically important deciduous tree species. Over the past few decades, increasing market demands and unprecedented human activity in its natural habitat have created new threats to this species. Nonetheless, the distribution of its habitat and the crucial environmental parameters that determine the habitat suitability remain largely unclear. The present study modeled the current and future geographical distribution of S. tzumu by maximum entropy (MAXENT) and genetic algorithm for rule set prediction (GARP). The value of area under the receiver operating characteristic curve (AUC), Kappa, and true skill statistic (TSS) of MAXENT was significantly higher than that of GARP, indicating that MAXENT performed better. Temperate and subtropical regions of eastern China where the species had been recorded was suitable for growth of S. tzumu. Relative humidity (26.2% of permutation importance), average temperature during the driest quarter (16.6%), annual precipitation (12.6%), and mean diurnal temperature range (10.3%) were identified as the primary factors that accounted for the present distribution of S. tzumu in China. Under the climate change scenario, both algorithms predicted that range of suitable habitat will expand geographically to northwest. Our results may be adopted for guiding the preservation of S. tzumu through identifying the habitats susceptible to climate change.

Exploring environmental coverages of species: a new variable contribution estimation methodology for rulesets from the genetic algorithm for rule-set prediction.

Variable contribution estimation for, and determination of variable importance within, ecological niche models (ENMs) remain an important area of research with continuing challenges. Most ENM algorithms provide normally exhaustive searches through variable space; however, selecting variables to include in models is a first challenge. The estimation of the explanatory power of variables and the selection of the most appropriate variable set within models can be a second challenge. Although some ENMs incorporate the variable selection rubric inside the algorithms, there is no integrated rubric to evaluate the variable importance in the Genetic Algorithm for Ruleset Production (GARP). Here, we designed a novel variable selection methodology based on the rulesets generated from a GARP experiment. The importance of the variables in a GARP experiment can be estimated based on the consideration of the prevalence of each environmental variable in the dominant presence rules of the best subset of models and its coverage. We tested the performance of this variable selection method based on simulated species with both weak and strong responses to simulated environmental covariates. The variable selection method generally performed well during the simulations with over 2/3 of the trials correctly identifying most covariates. We then predict the distribution of Toxostoma rufum (a bird with a cosmopolitan distribution) in the continental United States (US) and apply our variable selection procedure as a real-world example. We found that the distribution of T. rufum could be accurately modeled with 13 or 10 of 21 variables, using an UI cutoff of 0.5 or 0.25, respectively, arriving at parsimonious environmental coverages with good model accuracy. We also provide tools to simulate species distributions for testing ENM approaches using R.

Impact of Climate Change on the Distribution of Euscaphis japonica (Staphyleaceae) Trees

Analyzing the effects of climate change on forest ecosystems and individual species is of great significance for incorporating management responses to conservation policy development. Euscaphis japonica (Staphyleaceae), a small tree or deciduous shrub, is distributed among the open forests or mountainous valleys of Vietnam, Korea, Japan, and southern China. Meanwhile, it is also used as a medicinal and ornamental plant. Nonetheless, the extents of E. japonica forest have gradually shrunk as a result of deforestation, together with the regional influence of climate change. The present study employed two methods for modeling species distribution, Maxent and Genetic Algorithm for Rule-set Prediction (GARP), to model the potential distribution of this species and the effects of climate change on it. Our results suggest that both models performed favorably, but GARP outperformed Maxent for all performance metrics. The temperate and subtropical regions of eastern China where the species had been recorded was very suitable for E. japonica growth. Temperature and precipitation were two primary environmental factors affecting the distribution of E. japonica. Under climate change scenarios, the range of suitable habitats for E. japonica will expand geographically toward the north. Our findings may be used in several ways such as identifying currently undocumented locations of E. japonica, sites where it may occur in the future, or potential locations where the species could be introduced and so contribute to the conservation and management of this species.

Shifts in potential geographical distribution of Pterocarya stenoptera under climate change scenarios in China.

Climate change poses a serious threat to biodiversity. Predicting the effects of climate change on the distribution of a species' habitat can help humans address the potential threats which may change the scope and distribution of species. Pterocarya stenoptera is a common fast‐growing tree species often used in the ecological restoration of riverbanks and alpine forests in central and eastern China. Until now, the characteristics of the distribution of this species' habitat are poorly known as are the environmental factors that influence its preferred habitat. In the present study, the Maximum Entropy Modeling (Maxent) algorithm and the Genetic Algorithm for Ruleset Production (GARP) were used to establish the models for the potential distribution of this species by selecting 236 sites with known occurrences and 14 environmental variables. The results indicate that both models have good predictive power. Minimum temperature of coldest month (Bio6), mean temperature of warmest quarter (Bio10), annual precipitation (Bio12), and precipitation of driest month (Bio14) were important environmental variables influencing the prediction of the Maxent model. According to the models, the temperate and subtropical regions of eastern China had high environmental suitability for this species, where the species had been recorded. Under each climate change scenario, climatic suitability of the existing range of this species increased, and its climatic niche expanded geographically to the north and higher elevation. GARP predicted a more conservative expansion. The projected spatial and temporal patterns of P. stenoptera can provide reference for the development of forest management and protection strategies.

Characterization and risk assessment of the invasive papaya mealybug, Paracoccus marginatus, in Kenya under changing climate

AbstractThe present study was conducted to characterize the newly invasive papaya mealybug Paracoccus marginatus Williams and Granara de Willink in Kenya using molecular techniques and to establish the potential risk of spread of the pest. Although abundant literature of P. marginatus outbreaks exists in other parts of the world, studies from Africa are rare. Our results revealed significant similarity between Kenyan samples with GenBank accession number KP692333.1 of P. marginatus. Phylogenetic analyses generated a tree that was paraphyletic with two clusters showing low genetic distance values for the P. marginatus sequences, which diverged from that of Planococcus citri. Our models displayed an optimal performance with mean area under the curve value of 0.82 and 0.98 for Genetic Algorithm for Rule‐Set Production (GARP) and maximum entropy modelling (MaxEnt), respectively. Isothermality was the most influential variable in determining the potential distribution of P. marginatus with a 69% contribution to the models. Other variables included temperature mean diurnal range temperature seasonality, temperature annual range and annual precipitation in decreasing order of contribution. Current prediction by both models exceeded the existing range of P. marginatus, exacerbating the potential threat of the pest. GARP was more conservative in predicting suitable areas for P. marginatus, while MaxEnt showed further expansion by the year 2050. Our findings provide important information to guide biosecurity agencies in decision‐making to prevent the spread and enhance control efforts of P. marginatus.

A Modeling Comparison of Groundwater Potential Mapping in a Mountain Bedrock Aquifer: QUEST, GARP, and RF Models

In some arid regions, groundwater is the only source of water for human needs, so understanding groundwater potential is essential to ensure its sustainable use. In this study, three machine learning models (Genetic Algorithm for Rule-Set Production (GARP), Quick Unbiased Efficient Statistical Tree (QUEST), and Random Forest (RF)) were applied and verified for spatial prediction of groundwater in a mountain bedrock aquifer in Piranshahr Watershed, Iran. A spring location dataset consisting of 141 springs was prepared by field surveys, and from this three different sample datasets (S1–S3) were randomly generated (70% for training and 30% for validation). A total of 10 groundwater conditioning factors were prepared for modeling, namely slope percent, relative slope position (RSP), plan curvature, altitude, drainage density, slope aspect, topographic wetness index (TWI), terrain ruggedness index (TRI), land use, and lithology. The area under the receiver operating characteristic curve (AUC) and true skill statistic (TSS) were used to evaluate the accuracy of models. The results indicated that all models had excellent goodness-of-fit and predictive performance, but that RF (AUCmean = 0.995, TSSmean = 0.89) and GARP (AUCmean = 0.957, TSSmean = 0.82) outperformed QUEST (AUCmean = 0.949, TSSmean = 0.74). In robustness analysis, RF was slightly more sensitive than GARP and QUEST, making it necessary to consider several random partitioning options for preparing training and validation groups. The outcomes of this study can be useful in sustainable management of groundwater resources in the study region.

Assessing the Potential Distribution of Asian Gypsy Moth in Canada: A Comparison of Two Methodological Approaches

Gypsy moth (Lymantria dispar L.) is one of the world’s worst hardwood defoliating invasive alien species. It is currently spreading across North America, damaging forest ecosystems and posing a significant economic threat. Two subspecies L. d. asiatica and L. d. japonica, collectively referred to as Asian gypsy moth (AGM) are of special concern as they have traits that make them better invaders than their European counterpart (e.g. flight capability of females). We assessed the potential distribution of AGM in Canada using two presence-only species distribution models, Maximum Entropy (MaxEnt) and Genetic Algorithm for Rule-set Prediction (GARP). In addition, we mapped AGM potential future distribution under two climate change scenarios (A1B and A2) while implementing dispersal constraints using the cellular automation model MigClim. MaxEnt had higher AUC, pAUC and sensitivity scores (0.82/1.40/1.00) when compared to GARP (0.70/1.26/0.9), indicating better discrimination of suitable versus unsuitable areas for AGM. The models indicated that suitable conditions for AGM were present in the provinces of British Columbia, Ontario, Quebec, Nova Scotia and New Brunswick. The human influence index was the variable found to contribute the most in predicting the distribution of AGM. These model results can be used to identify areas at risk for this pest, to inform strategic and tactical pest management decisions.

Prediction of potential distribution of the invasive species Procambarus clarkii in China based on ecological niche models

Procambarus clarkii was introduced into China as an important aquatic product in early 20th century. It has characteristics of high fertility, rapid growth, adaptability and digging burrows, which could cause damage of crops, cropland and facilities, decrease local biodiversity and thus threaten local ecosystem. Thus, predicting the potential distribution of P. clarkii in response to climate change was essential for preventing and monitoring this species. Based on the distribution of P. clarkii, the maximum entropy (MaxEnt) and genetic algorithm for rule-set production (GARP) models were used to predict its distribution in China under current climate and four climate scenarios (RCP 2.6, RCP 4.5, RCP 6.0, RCP 8.5) in two periods, 2041-2060 and 2061-2080. Then, the modeling results were tested by ROC curves. The results showed that under current climate, the highly suitable region for distribution predicted by the MaxEnt and GARP models were Shanghai, Jiangsu, Zhejiang and Anhui along the Yangtze River. The main environmental variables affecting its distribution were mean temperature of the coldest quarter, minimum temperature of the warmest month, and temperature seasonality, maximum temperature of the warmest month, precipitation of the driest month. Under the future climate scenarios, the suitable area of P. clarkii distribution varied in 2061-2080. The total suitable area of P. clarkii would increase under RCP2.6 and RCP 4.5, whereas under RCP 8.5 the suitable area of P. clarkii would increase, and then decrease. In RCP 6.0, there was no change. The suitable areas of P. clarkii would disperse to different latitude areas and migrate toward high altitude.

Identification of high-risk habitats of Oncomelania hupensis, the intermediate host of schistosoma japonium in the Poyang Lake region, China: A spatial and ecological analysis.

BackgroundIdentifying and eliminating snail habitats is the key measure for schistosomiasis control, critical for the nationwide strategy of eliminating schistosomiasis in China. Here, our aim was to construct a new analytical framework to predict high-risk snail habitats based on a large sample field survey for Oncomelania hupensis, providing guidance for schistosomiasis control and prevention.Methodology/Principal findingsTen ecological models were constructed based on the occurrence data of Oncomelania hupensis and a range of variables in the Poyang Lake region of China, including four presence-only models (Maximum Entropy Models, Genetic Algorithm for rule-set Production, Bioclim and Domain) and six presence-absence models (Generalized Linear Models, Multivariate Adaptive Regression Splines, Flexible Discriminant Analysis, as well as machine algorithmic models–Random Forest, Classification Tree Analysis, Generalized Boosted Model), to predict high-risk snail habitats. Based on overall predictive performance, we found Presence-absence models outperformed the presence-only models and the models based on machine learning algorithms of classification trees showed the highest accuracy. The highest risk was located in the watershed of the River Fu in Yugan County, as well as the watershed of the River Gan and the River Xiu in Xingzi County, covering an area of 52.3 km2. The other high-risk areas for both snail habitats and schistosomiasis were mainly concentrated at the confluence of Poyang Lake and its five main tributaries.Conclusions/SignificanceThis study developed a new distribution map of snail habitats in the Poyang Lake region, and demonstrated the critical role of ecological models in risk assessment to directing local field investigation of Oncomelania hupensis. Moreover, this study could also contribute to the development of effective strategies to prevent further spread of schistosomiasis from endemic areas to non-endemic areas.