Advanced Search

Preplanned Studies: Modelling the Ecological Suitability of the Chagas Disease Vector Triatoma rubrofasciata — China

View author affiliations
  • Summary

    What is already known on this topic?

    Triatoma rubrofasciata is a potential vector that can transmit American trypanosomiasis and was widely recorded in South of China.

    What is added by this report?

    Because of the low density of the triatomines, more habitats have not been discovered. This study mainly focused on predicting the geographical distribution of T. rubrofasciata under current and future climatic conditions in China using the MaxEnt model.

    What are the implications for public health practice?

    The result showed that the distribution of T. rubrofasciata was largely affected by annual mean temperature and possessed a high potential for expansion in southern China in the future. Our predictions are useful for targeting surveillance efforts in high-risk areas and increasing the efficiency and accuracy of public health investigations and vector control efforts in China.

  • loading...
  • Funding: Supported by National Science and Technology Project (No.2018ZX10101002), National Key Research and Development Program of China (Grant No.2016YFC1202000), and Guangxi Natural Science Foundation Program (2018GXNSFAA294092)
  • [1] Liu Q, Zhou XN. Preventing the transmission of American trypanosomiasis and its spread into non-endemic countries. Infect Dis Poverty 2015;4:60. http://dx.doi.org/10.1186/s40249-015-0092-7CrossRef
    [2] Pérez-Molina JA, Molina I. Chagas disease. Lancet 2018;391(10115):82 − 94. http://dx.doi.org/10.1016/S0140-6736(17)31612-4CrossRef
    [3] Liu Q, Wu FW, Shi YL, Lu WC, Xie HG, Guo YH, et al. Distribution of triatomines, the vector of chagas disease — southern China, 2016−2018. China CDC Wkly 2020;2(33):629 − 33. http://dx.doi.org/10.46234/ccdcw2020.174CrossRef
    [4] Phillips SJ, Anderson RP, Schapire RE. Maximum entropy modeling of species geographic distributions. Ecol Model 2006;190(3 − 4):231 − 59. http://dx.doi.org/10.1016/j.ecolmodel.2005.03.026CrossRef
    [5] Richman R, Diallo D, Diallo M, Sall AA, Faye O, Diagne CT, et al. Ecological niche modeling of Aedes mosquito vectors of chikungunya virus in southeastern Senegal. Parasit Vectors 2018;11(1):255. http://dx.doi.org/10.1186/s13071-018-2832-6CrossRef
    [6] Sage KM, Johnson TL, Teglas MB, Nieto NC, Schwan TG. Ecological niche modeling and distribution of Ornithodoros hermsi associated with tick-borne relapsing fever in western North America. PLoS Negl Trop Dis 2017;11(10):e0006047. http://dx.doi.org/10.1371/journal.pntd.0006047CrossRef
    [7] Artun O. Ecological niche modeling for the prediction of cutaneous leishmaniasis epidemiology in current and projected future in Adana, Turkey. J Vector Borne Dis 2019;56(2):127 − 33. http://dx.doi.org/10.4103/0972-9062.263726CrossRef
    [8] Shi YL, Wei YB, Feng XY, Liu JF, Jiang ZH, Ou FQ, et al. Distribution, genetic characteristics and public health implications of Triatoma rubrofasciata, the vector of Chagas disease in Guangxi, China. Parasit Vectors 2020;13(1):33. http://dx.doi.org/10.1186/s13071-020-3903-zCrossRef
    [9] Chen MX. Prevention and treatment of the dermatitis caused by Triatoma. Hainan Med 1986;1:35.
    [10] Zhang G. Study on the medical significance of the Triatoma. Hainan Med 1991;4:8 − 9.
    [11] Wang Y, Peng L. Four cases of anaphylactic shocks caused by triatomine bugs. Clin Focus 2006;21:1059 − 61.
    [12] Eberhard FE, Cunze S, Kochmann J, Klimpel S. Modelling the climatic suitability of Chagas disease vectors on a global scale. Elife 2020;9:e52072. http://dx.doi.org/10.7554/eLife.52072CrossRef
    [13] Ceccarelli S, Balsalobre A, Susevich ML, Echeverria MG, Gorla DE, Marti GA. Modelling the potential geographic distribution of triatomines infected by Triatoma virus in the southern cone of south America. Parasit Vectors 2015;8:153. http://dx.doi.org/10.1186/s13071-015-0761-1CrossRef
    [14] IPCC. Climate change 2001: the scientific basis. Contribution of working group I to the third assessment report of the intergovernmental panel on climate change. Cambridge, United Kingdom and New York, NY, USA: Cambridge University Press. 2001. https://www.oalib.com/references/11734021.https://www.oalib.com/references/11734021
    [15] Vollering J, Halvorsen R, Mazzoni S. The MIAmaxent R package: variable transformation and model selection for species distribution models. Ecol Evol 2019;9(21):12051 − 68. http://dx.doi.org/10.1002/ece3.5654CrossRef
  • FIGURE 1.  Relative predictive power of different environmental variables based on the jackknife of regularized training gain in maximum entry modeling for T. rubrofasciata in China.

    FIGURE 2.  Distribution maps of T. rubrofasciata under current and future climatic change condition (2050 and 2070) based on A2a/HadCM3 scenario.

    FIGURE 3.  Distribution maps of T. rubrofasciata under future climatic change condition (2050 and 2070) based on binary species distribution modelling (SDMs).

Citation:

通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索
Turn off MathJax
Article Contents

Article Metrics

Article views(7425) PDF downloads(41) Cited by()

Share

Related

Modelling the Ecological Suitability of the Chagas Disease Vector Triatoma rubrofasciata — China

View author affiliations

Summary

What is already known on this topic?

Triatoma rubrofasciata is a potential vector that can transmit American trypanosomiasis and was widely recorded in South of China.

What is added by this report?

Because of the low density of the triatomines, more habitats have not been discovered. This study mainly focused on predicting the geographical distribution of T. rubrofasciata under current and future climatic conditions in China using the MaxEnt model.

What are the implications for public health practice?

The result showed that the distribution of T. rubrofasciata was largely affected by annual mean temperature and possessed a high potential for expansion in southern China in the future. Our predictions are useful for targeting surveillance efforts in high-risk areas and increasing the efficiency and accuracy of public health investigations and vector control efforts in China.

  • 1. National Institute of Parasitic Diseases, Chinese Center for Disease Control and Prevention (Chinese Center for Tropical Diseases Research); NHC Key Laboratory of Parasite and Vector Biology; WHO Collaborating Centre for Tropical Diseases; National Center for International Research on Tropical Diseases, Shanghai, China
  • 2. Institute of Parasitic Disease Prevention and Control, Guangxi Zhuang Autonomous Region Center for Disease Control and Prevention, Nanning, Guangxi Zhuang Autonomous Region, China
  • Corresponding author:

    Qin Liu, liuqin@nipd.chinacdc.cn

  • Funding: Supported by National Science and Technology Project (No.2018ZX10101002), National Key Research and Development Program of China (Grant No.2016YFC1202000), and Guangxi Natural Science Foundation Program (2018GXNSFAA294092)
  • Online Date: August 27 2021
    Issue Date: August 27 2021
    doi: 10.46234/ccdcw2021.182
  • Chagas disease, caused by Trypanosoma cruzi, is considered the fourth most transmitted disease after malaria, tuberculosis, and schistosomiasis by World Health Organization (WHO) (1). Chagas disease has become a global health issue in recent decades due to the spreading worldwide (2). Although there has not yet been any recorded cases of Chagas disease in China, one of the potential vectors, Triatoma rubrofasciata that can transmit T. cruzi, was widely recorded in southern China (3). In this study, the maximum entropy method (MaxEnt) for species distribution modeling (SDM) was used to predict the geographical distribution of T. rubrofasciata, under current and future climatic conditions (BCC-CSM-1 RCP26) in China. A total of 184 locations with the species occurrence were recorded; 19 bioclimatic variables derived from monthly data were used in the modeling process. The results showed that the distribution of T. rubrofasciata was largely affected by annual mean temperature and minimum temperature of the coldest month. According to species response curves, this species preferred habitats with annual mean temperatures of 22.5 °C. Climate change projections indicated that by 2050 and 2070, the habitats of the T. rubrofasciata would be expanded.

    T. rubrofasciata is one of the potential vectors of Chagas disease that is recorded to be the most widely distributed worldwide. In China, we started monitoring triatomines and imported Chagas disease in 2016. During a survey from 2016–2018, 170 habitats were recorded (3), but because of low density of the triatomines, many habitats were likely underrepresented. Therefore, predicting these habitats through appropriate environmental data is important to make up for this underrepresentation.

    MaxEnt is a machine learning model that estimates a target probability distribution by calculating the probability distribution of maximum entropy (4). MaxEnt has been used extensively to model the distribution of several vector-reliant disease-causing pathogens (5-6). Many scholars believed that the MaxEnt model was better than other models in predicting the spatial distribution of species, especially in cases of incomplete distribution data (6-7). MaxEnt is also capable of projecting shifts in species distribution under various climate change scenarios (4). In this study, MaxEnt software was used to predict the potential distribution of T. rubrofasciata. A total of 184 records have been collected in the database. These came from two sources: 1) published records on T. rubrofasciata in journals, books, and reports and from Shanghai Insect Museum and Museum of insects Institute of Guangdong Province in China (8-11); and 2) specimen field surveys by the authors from 2016 (3). After screening duplicates and poor data, 116 records were obtained for the construction of the final model.

    For present environmental data (1950–2000), a total of 19 bioclimatic layers were downloaded from Worldclim database (www.worldclim.org) (Supplementary Table S1). All data used for SDM had a spatial resolution of 1 km2 (30 arc seconds). The data of distribution points and environmental variables were inputted into MaxEnt software. The maximum number of iterations was set to 2000, bootstrap repeated 10 times, and random seed was turned on. Then, the response curves and jackknife test functions were checked to analyze the importance of environmental variables that affected the distribution of T. rubrofasciata. The establishment of the model needs to randomly select 75% of the distribution points, and the test of the model is completed by data of the remaining 25% of the occurrence points to construct an optimal habitat model of T. rubrofasciata. The prediction results of MaxEnt model were tested by the area under the receiver operating characteristic curve (ROC). Maximum entropy modeling was used with the MaxEnt software (version 3.3.3k, American Museum of Natural History, NY, USA) for quantifying relative risk of invasion and mapping of the potential geographic distribution of T. rubrofasciata in China. A jackknife procedure was used to calculate the significance of the contribution of each variable to the model. The area under the receiver operating characteristic curve [area under curve (AUC)] was used to evaluate model performance. The potential range of changes in the suitable areas under different scenarios of climate change in the future was analyzed by using the “distribution changes between binary SDMs” tool. When performing the quick recursive to binary conversion, the threshold value was 0.5. ArcMap was loaded for all maps..

    The MaxEnt model prediction has high AUC (0.989) indicating good model performance for T. rubrofasciata (Supplementary Figure S1). The jackknife test showed that the distribution of T. rubrofasciata was largely affected by annual mean temperature (Bio1), mean temperature of the coldest quarter (Bio11), and minimum temperature of coldest month (Bio6) (Figure 1). Annual mean temperature (Bio1) explained 63.4% of the total variance and was thus identified as the main factor affecting the spatial distribution of T. rubrofasciata. Species distribution maps showed that Guangdong, Guangxi, Fujian, and Taiwan, China of the study area were recognized as high potential habitats of T. rubrofasciata in the present conditions, by 2050, and by 2070 (Figure 2). The averaged future predictions of MaxEnt for 2050 and 2070 (binary SDMs) revealed an expansion in suitable habitats of T. rubrofasciata of 180,433 km2 and 167,495 km2 respectively, which reached more northern latitudes of southern China (Figure 3).

    Figure 1. 

    Relative predictive power of different environmental variables based on the jackknife of regularized training gain in maximum entry modeling for T. rubrofasciata in China.

    Figure 2. 

    Distribution maps of T. rubrofasciata under current and future climatic change condition (2050 and 2070) based on A2a/HadCM3 scenario.

    Figure 3. 

    Distribution maps of T. rubrofasciata under future climatic change condition (2050 and 2070) based on binary species distribution modelling (SDMs).

  • T. rubrofasciata, which was recorded as a potential vector that can transmit American trypanosomiasis, was the most widely distributed worldwide. During the survey of 2016–2018, the habitats of T. rubrofasciata was widely recorded in southern China, as well as exhibiting domestic and peri-domestic behavior invading chicken coops and human dwellings. However, due to its low density, more habitats would likely be underrepresented, including, for example, historical reports in Yunnan Province that were not corroborated in current surveys (2). For this purpose, we used MaxEnt to identify regions that offer climatically suitable conditions for this species.

    The discriminatory capacity of the model displayed good predictive performance, which was also reflected in the AUC values of over 0.95. Studies have indicated that potential hotspots of triatomine species diversity in tropical and subtropical regions between 21˚N and 24˚S latitude and a major limiting factor of triatomine distribution was temperature seasonality (12-13). In our study, after removing auto-correlated parameters (Figure 1), MaxEnt indicated that the current distribution of T. rubrofasciata was more affected by temperature variables (Bio1, Bio6, and Bio11). Among them, Bio1 (annual mean temperature) was the most affected and explained 63.4% of the total variance. The International Panel of Climate Change (IPCC) has predicted that annual mean temperatures will increase up to 5.8 °C by the end of this century (14). As the global climate warms, the risk of T. rubrofasciata expansion may increase.

    The information obtained from the predictive maps of the current distribution of T. rubrofasciata was compared to those assembled from the series of future predictions in 2050 and 2070 with a medium GHG concentration scenario (RCP26). Global climate models trained on the existing potential distribution showed a relatively stable estimate for the amount of land area that was classified as suitable for T. rubrofasciata. The two emissions scenarios we modeled (RCP 26) produced very similar predicted distributions, although the pathway of high concentrations of GHG predicted slightly less overall area (Figure 2). There was a predicted shift in the distribution with suitable areas moving from lower latitude and presumably warmer climates to climates at higher latitude where conditions may become more suitable (Figure 2). Regions of high predicted probability in 2050 were found in Guangxi, Guangdong and Hainan, and good suitable areas were expanded in Southern of Fujian, Jiangxi, Hunan, Guizhou, and Yunnan.

    In general, using this model, it was easy to find climatically suitable habitats of T. rubrofasciata in China. It was useful for targeting surveillance and increasing the efficiency and accuracy of public health investigations and vector control efforts in China in the future. Background surveys should focus on southern China and set up more sampling survey points, while a small number of sampling survey points could be set up in the southern areas of Yunnan, Guizhou, Jiangxi, and Fujian.

    However, the present study has some limitations. Although more than 100 records were obtained for the construction the model, the data was mainly concentrated in Guangdong and Guangxi, so its representation has certain limitations. More data would be collected to add to the model to get more accurate prediction. The widely used “MaxEent” software for modeling species distributions from presence-only data tends to produce models with high-predictive performance but low-ecological interpretability, which are more complex but not necessarily predictively better-than subset selection. The simpler models implement two different kinds of model fitting: maximum entropy fitting for presence-only data and logistic regression for presence-absence data would be new motivation field. With improvement of the model, some model which could cover the shortage of the data, eg, MIAmaxent R package etc, which would be involved in the habitats of T. rubrofasciata prediction in future study (15).

    Conflicts of Interest: No conflicts of interest were reported.

    Acknowledgments: All participants of this surveillance program.

Reference (15)

Citation:

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return