Next Article in Journal
Knowledge of Trainee Teachers about the New Water Culture: A Comparative Study between Two Spanish Cities
Previous Article in Journal
Learning Analytics in Supporting Student Agency: A Systematic Review
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sustainability
Volume 15
Issue 18
10.3390/su151813669
sustainability-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Predicting the Spatial Distribution of Hyalomma ssp., Vector Ticks of Crimean–Congo Haemorrhagic Fever in Iraq

by
Nabaz R. Khwarahm
Department of Biology, College of Education, University of Sulaimani, Sulaimani 334, Kurdistan Region, Iraq
Sustainability 2023, 15(18), 13669; https://doi.org/10.3390/su151813669
Submission received: 5 August 2023 / Revised: 3 September 2023 / Accepted: 11 September 2023 / Published: 13 September 2023
(This article belongs to the Section Health, Well-Being and Sustainability)
Browse Figures
Versions Notes

Abstract

:
Crimean–Congo hemorrhagic fever (CCHF) typically spreads through ticks and is categorized as a viral hemorrhagic fever. CCHF is a fatal endemic disease in Iraq, and it has been reported sporadically since its first report in 1979. Recent outbreaks during 2021–2023 and their fatal consequences captured the interest of this study. CCHF is a tick-borne disease that represents a major challenge to the public health, social, and economic sectors. The geographical distribution of CCHF is closely linked with Hyalomma vector tick distribution. Therefore, predicting and mapping the spatial distribution of the disease vector in relation to relevant environmental factors provides invaluable information for establishing an early warning system based on which preventive measures can be taken to minimize the spread and, hence, the fatal consequences of CCHF. To achieve this, this study incorporates geospatial techniques and maximum entropy modeling (Maxent) to assess the habitat suitability of the Hyalomma vector and to identify the key environmental drivers contributing to its spatial distribution in Iraq. Utilizing the area under the ROC curve (AUC) as the performance metric, the model evaluation yielded successful results in predicting habitat suitability for Hyalomma vector ticks in Iraq. The AUC attained an average score of 0.885 with a regularization multiplier (β) set at 1. The Hyalomma ticks’ suitable habitat distribution within the study area covers a fraction of the total land, at approximately 51% (225,665 km2) of the entire 441,724 km2 region. Among these suitable areas, 41.57% (183,631 km2) were classified as lowly suitable, 8.61% (38,039 km2) as moderately suitable, and 0.9% (3994 km2) as highly suitable. Several factors have significantly influenced Hyalomma vector tick distribution in Iraq. These include land cover (accounting for 50.8%), elevation (contributing 30.4%), NDVI (5.7%), temperature seasonality (4.7%), precipitation seasonality (3.3%), sheep density (2.3%), goat density (2.2%), and the mean diurnal range (0.5%). The findings of this study could have significant implications for establishing a strategic early warning system and taking preventive measures beforehand to minimize and control Crimean–Congo haemorrhagic fever in Iraq and similar ecoregions in the Middle East. As a primary precaution, this study recommends focusing on highly suitable areas (3994 km2) in the southern part of Iraq for management and preventive actions.

1. Introduction

Crimean–Congo hemorrhagic fever (CCHF) is an endemic disease in Iraq, with intermittent documented instances tracing back to its initial appearance in the 1970s [1]. CCHF is a viral disease transmitted by ticks, posing significant challenges to public health as well as the social and economic sectors. The geographical spread of CCHF largely coincides with the distribution of the Hyalomma vector ticks [2,3]. Apart from Iraq, CCHF also occurs in parts of Africa, Asia, the Middle East, and southeastern Europe [4]. People acquire the CCHF virus through the bite of Hyalomma ticks that are infected, direct contact with the blood or bodily fluids of individuals experiencing the acute phase of the disease, and by coming into contact with the blood or tissue of livestock that carry the virus [5].
The initial outbreak of CCHF was historically identified in the Crimean area of the former Soviet Union back in 1944 [4]. Subsequently, the virus spread to various regions, such as Bulgaria, China, the former Yugoslavia, Pakistan, the United Arab Emirates, Iraq, and parts of Russia [2]. Later on, CCHF was also reported in the Balkan countries, southwest Russia, the Middle East, India, Turkey, and southwestern Europe [6]. The expansion of CCHF’s geographic range has been attributed to climate change factors, migratory birds (transporting ticks), and animal trade [7]. As such, understanding the contributing factors to the distribution of tick species can provide useful information on the possible emergence of CCHF in new geographical ranges with the changing environment (climate change and anthropogenic-related factors). Previous studies have explored several contributing factors that facilitate the spatial distribution of Hyalomma ticks at the regional level. For example, in Tajikistan, Central Asia, the Balkans, and the Anatolian Peninsula, various environmental factors, including climatic factors, topographic factors, and livestock density, were considered in assessing the spatial distribution of CCHF disease ecology by employing the species distribution modeling (SDM) approach [8,9,10,11]. Furthermore, another study modeled eight European tick species in the western Palearctic region using SDMs and compared them with null models [12]. In the USA, the Maxent modeling approach was used to predict the habitat distribution pattern of Rhipicephalus sanguineus sensu lato ticks [13]. Apart from CCHF, other infectious diseases, such as Keratoconjunctivitis in Chamois, Bluetongue vectors, West Nile disease, and Rift Valley fever, were also explored using environmental variables (including climate change conditions) and disease prevalence and geographical distribution, respectively [14,15,16].
In Iraq, studies on CCHF disease are limited to case reports, transmission, and control measures [3,17]; epidemiological characteristics of CCHF cases [1]; trend analysis of CCHF cases between 1990 and 2010 [18]; isolation and identification [19]; and seroepidemiological surveys [20]. Recent outbreaks in 2021, 2022, and 2013 captured the interest of the current investigation into distribution of the disease. For example, the Ministry of Health in Iraq has reported a significant increase in cases during the past two years. In 2021, there were 19 confirmed cases based on laboratory data, and the number rose to 108 laboratory-confirmed cases in the first half of 2022 [1]. In 2023, according to local media news (as of 13 July 2023) from the Minister of Health, since January, 375 confirmed cases have been recorded across the country. The steady increase in reported cases across the country, mainly in the southern parts, and its distribution to the northern provinces, raise the question of the possible factors and ecological niche requirements of the vector ticks in Iraq. As far as this study is aware, no studies in Iraq have been conducted on predicting and mapping the geographical distribution of the key vector ticks responsible for the transmission of CCHF disease in Iraq. To fill in this research gap and establish new baseline information on the environmental requirements of the vector ticks in Iraq and the Middle East, this study was conceptualized.
Examining and comprehending the spatial distribution of Hyalomma vector ticks offer indispensable foundational data for the implementation of precautionary and preventive measures before the outbreak of the CCHF disease. Therefore, accurately predicting and mapping the spatial distribution of the disease vector, while taking relevant environmental factors into account, offers crucial insights for establishing an early warning system. This system can help implement preventive measures to minimize the spread and mitigate the fatal consequences of CCHF. To achieve this, this study incorporates geospatial techniques and maximum entropy modeling (Maxent) to assess the suitability of the Hyalomma vector’s habitat and to identify the key environmental factors contributing to its distribution in Iraq.

2. Materials and Methods

2.1. Study Area

Iraq, which spans a total area of approximately 441,724 km2 and is situated in the Middle East, occupies a geographical position between latitudes 29° and 38° N and longitudes 39° and 49° E. It shares borders with Iran, Kuwait, Saudi Arabia, Jordan, Syria, and Turkey. From a geographic perspective, Iraq can be categorized into four distinct regions [21]. The northeastern part consists of mountainous terrain, the southern area comprises marshlands, the western region is primarily desert, and the uplands serve as a transitional zone between the desert and mountain ranges [21]. The climate in Iraq varies across these regions. The western and southern parts of the country experience predominantly dry weather, while the central and southern regions have a range of climates, ranging from continental to arid. The northern and northeastern parts, which are characterized by mountains, have a Mediterranean climate. During winter, the average temperature is around 16 °C. However, the summer season is exceptionally hot, with daytime temperatures soaring above 43 °C in July and August but dropping to around 26 °C at night. Precipitation in Iraq typically occurs from December to February, although in the northern and northeastern regions, rainfall can extend from November to April [22] (Figure 1).

2.2. Occurrence Data

The occurrence points of vector ticks, Hyalomma genera, were acquired from the Biodiversity Information Facility (GBIF) website (https://www.gbif.org/occurrence/download/0045370-230530130749713 (accessed on 1 April 2023)). The occurrence records were assessed for their quality, including the accuracy of their positions, by overlaying them within a GIS environment. Initially, there were 61 occurrence points gathered from the GBIF. However, after implementing spatial filtering [23] and removing duplicate records, the number of points was reduced to 11 for model building and analysis. Spatial filtering reduces sampling bias by allowing a reasonable distance between the occurrence points. The model choice in this study was the maximum entropy model (Maxent), which is known to be insensitive to small sample sizes [24,25].

2.3. Environmental Predictors

The model’s development relied on various environmental predictors (Table 1 and Table 2). These predictors were selected based on the available literature [9,10] and local knowledge of the disease prevalence across Iraq. Climatic predictors, representing 19 bioclimatic variables for current conditions, were obtained from the website (www.worldclim.org (accessed on 1 February 2022)), a dataset published by the Intergovernmental Panel on Climate Change (IPCC) in Assessment Report 5 (AR5). The WorldClim dataset comprises 19 bioclimatic variables that were initially calculated based on monthly temperature and rainfall data collected from weather stations during the period from 1950 to 2000 [26]. These data were obtained at a spatial resolution of 30 s, which is approximately equivalent to 1 km. These datasets are available on a global scale. Spatial analyst tools, such as ‘extract by mask’, were utilized to scale up to the extent of the study area. The same procedure was applied to other environmental layers.
The topographic predictor, the Digital Elevation Model (DEM), was sourced from the Shuttle Radar Topography Mission (SRTM) dataset provided by the Consultative Group for International Agricultural Research (CGIAR) Consortium (http://srtm.csi.cgiar.org/-srtmdata/ (accessed on 1 February 2022)). Additionally, livestock density predictors, such as sheep, goats, cattle, chickens, camels, and buffalos, which are available at a global scale with 5 min of arc ground resolution, were acquired from (www.fao.org/geonetwork/srv/en/main.home (accessed on 1 March 2022)). These datasets represent the absolute number of animals per pixel for 2015 according to the Gridded Livestock of the World database (v4) [27]. Land cover was another predictor considered in model building in this study. For this, a recent global land cover map for 2019 [28] was acquired from (https://zenodo.org/communities/copernicus-land-cover/?page=1&size=20 (accessed on 1 February 2022)). This product has a 100 m spatial resolution.
The Normalized Difference Vegetation Index (NDVI), which serves as an important primary producer and, hence, a surrogate for vegetation stock, for the growing season of March to September 2021 was generated from Landsat 8 imagery. These datasets were downloaded from (https://earthexplorer.usgs.gov (accessed on 1 February 2022)) at 30 m spatial resolution. The NDVI was derived using the formula NDVI = (Nir − Red)/(Nir + Red), with Landsat 8 bands Nir = Band 5 and Red = Band 4. To ensure comparability among the predictors, all datasets were rescaled to approximately 1 km using ArcGIS 10.3 software. Collinearity is known to impact the accuracy of correlative-based modeling techniques; therefore, before model construction, a threshold technique was employed, retaining only those with a Pearson’s pairwise correlation coefficient of |r| ≤ 0.8 [29] (Table 1). Only the predictors listed in Table 2 were included in the model building; due to their high correlation issues, other predictors, such as cattle, chickens, camels, buffalos, and the majority of the bioclimatic predictors, were eliminated (Table 2).

2.4. Model Building

This study’s model selection relied on predictive capability and occurrence data as the primary criteria. Therefore, the chosen model for this research was the Maxent model, which is based on machine learning techniques, owing to its reputation for accurate predictions [30,31]. The fundamental idea behind the Maxent model is to choose a probability distribution that maximizes information content and minimizes assumptions, while aligning with the given observed data. This strategy guarantees a fair and impartial portrayal of uncertainty, especially in scenarios with restricted information. Notably, Maxent exhibits robustness towards small sample sizes, requiring as few as 4–5 presence records to perform effectively [25].
The model creation involved a random partition of 80% of the presence data points for training purposes, while the remaining 20% were utilized for model validation [32]. Often, researchers tend to use default-mode configurations for model settings, which are anticipated to produce satisfactory results [33]. Nevertheless, it is crucial to take into account the nature of the input data, the modeler’s familiarity with the study area, and the specific configuration of the model, as these factors can have a substantial influence on the accuracy of the model’s predictions. Morales et al. [34] advises against relying solely on default settings, as they may not ensure dependable and trustworthy outcomes. This study employed 10 replicates of a model, employing the maximum entropy algorithm with 500 iterations. The outcome produced average probability maps for the occurrence of the victor ticks based on these 10 model replicates. To accommodate the number of the occurrence records, 9000 background points were used [35]. The regularization multiplier (β) in the model was set to the default value of one, as recommended by Merow et al. [36].
The Leave-One-Out cross-validation method (Jackknife test) was employed to evaluate the variables’ significance and their influence on the vector tick habitat distribution probability. This technique assesses the robustness of the model’s predictions. In addition, it helps evaluate the model’s performance by iteratively leaving out one occurrence point at a time and then comparing the model’s predictions with the omitted occurrence point. Additionally, the logistic output format was used to create a continuous map, employing the minimum training presence logistic threshold [37]. Based on this threshold, the habitat suitability and unsuitability for the vector ticks were categorized. Then, the suitability maps were reclassified into the following classes:
(i) unsuitable (0.0–0.20); (ii) low suitable (0.20–0.33); medium suitable (0.33–0.43); and high suitable (0.43–0.89) [38]. To perform these categorizations, spatial analysis tools within the ArcGIS 10.3 platform were utilized [38,39]. The unsuitable class signifies areas with the lowest probability of encountering the victor ticks and, hence, CCHF. On the other hand, the suitable classes represent areas with the highest probability of finding the species.

2.5. Model Evaluation

This study utilized the extensively acknowledged evaluation metric known as the Area Under Curve (AUC) (Hanley and McNeil 1982) to evaluate the model’s performance. The AUC values act as a gauge of discrimination strength, ranging from zero to one. Values approaching 1 signify exceptional discrimination, while a value of 0.5 suggests randomness in presence, and values above 0.5 to 1 indicate a gradual distinction between suitable and unsuitable areas for Hyalomma vector ticks (i.e., the probability of occurrence) [31].

3. Results

3.1. Model Performance

The model evaluation metric, Area Under the ROC Curve (AUC), yielded successful results in predicting habitat suitability and unsuitability for Hyalomma vector ticks across the study area. AUC scored an average of 0.885 with a standard deviation of ±0.026 (Figure 2). These values represent an average over the model replicate runs (i.e., 10 replicates).

3.2. Factors Contributing to Habitat Suitability Distribution of Hyalomma Vector Ticks

The distribution of the Hyalomma vector tick in Iraq is significantly influenced by various factors. Land cover has the most significant impact at 50.8%, followed by elevation at 30.4%, NDVI at 5.7%, temperature seasonality at 4.7%, and sheep and goat density at 2.3% and 2.2%, respectively. Additionally, precipitation seasonality and the mean diurnal temperature range also played roles in influencing the habitat distribution of Hyalomma vector ticks (Table 2). Similar patterns of variable contributions to the distribution of Hyalomma vector ticks were also demonstrated by the gains achieved through regularized training of the Jackknife test (Figure 3). For example, land cover and elevation demonstrated significant gains, among other variables. Meanwhile, the mean diurnal range (max temp–min temp) demonstrated the least gain.

3.3. Habitat Suitability Distribution of Hyalomma Vector Ticks

In Iraq, the species’ suitable habitat is limited to a portion of the total land, representing approximately 51% (225,665 km2) of the entire 441,724 km2 region. Among these suitable areas, the majority, 41.57% (183,631 km2), were considered to have low suitability, while 8.61% (38,039 km2) were moderately suitable, and only 0.9% (3994 km2) were highly suitable. High-risk areas for the spread of CCHF include the southern parts of the Dhi-Qar, Mysan, and Basra provinces. Parts of the capital city, Baghdad, are also risky areas for the dispersal of the vector ticks under the conditions considered in the model building (Figure 4). Moderately risky areas include certain sites across Diyala, Wasit, and Almuthanna in the southern part of Iraq. In the north, small areas in Duhok, Erbil, Kirkuk, and Ninaw also fall within the moderately risky areas (Figure 4). Significant portions of the country, such as the north, northeast, and western areas of Anbar, demonstrated unsuitable habitats for Hyalomma vector ticks in Iraq (Table 3).

4. Discussion

4.1. Modeling Limitations

The process of making predictions using models, whether individual or combined (ensemble models), involves certain inherent uncertainties. These uncertainties are influenced by various factors, such as biological aspects, the quality and consistency of input data, and climate models [40]. Certain measures are usually taken to minimize uncertainties, like verifying spatial accuracy, spatial thinning, and creating bias files during model building; the model outputs can still have some level of uncertainty, particularly concerning the limited number of occurrence records in this study. Although Maxent is known for working well under such circumstances, the spatial representation of occurrence records within the model boundary may produce outputs that are more reliable. Previous studies have already highlighted the presence of uncertainties in different species distribution models (SDMs) when estimating the spatial representation of species [41,42]. In this research, necessary steps were taken to minimize potential uncertainties, including checking spatial accuracy and ensuring a representative spatial distribution of occurrence records. Although models are chosen based solely on performance metrics, they may not fully represent the true uncertainties associated with the predictions. Understanding the contribution of each model to an averaged prediction can be a complex task, but accurately quantifying and managing uncertainty continue to be challenges in distribution modeling.
Modeling in this study neglected the significance of data related to wild animals (e.g., wild boar and birds) in influencing the potential distribution of Hyalomma ticks in environmentally suitable areas. Nevertheless, despite this absence of data, this study effectively mapped the environmental suitability of Hyalomma ticks capable of transmitting the CCHF virus. Moreover, this study marked the first attempt to estimate the ecological niches of Hyalomma ticks in Iraq. Therefore, further studies should consider taking into account various wild animals, such as pigeons, hedgehogs, foxes, dogs, and jackals, confirmed case reports of CCHF, and, most importantly, tick species profiles, particularly hard ticks, as they are usually associated with the CCHF disease. In addition, particular attention should be given to establishing spatial information (tracking data) on herd traders at both local and regional levels.

4.2. Contributing Factors to the Habitat Suitability Distribution of Hyalomma Vector Ticks

Having insights into the possible habitat of ticks is essential for anticipating the occurrence and recurrence of tick-borne diseases [43]. The interplay between ticks and the climate is a significant determinant, as the well-being and functions of ticks hinge upon external micro-abiotic conditions [44]. Moreover, the presence of appropriate hosts profoundly influences the dispersion of tick species within a specific region. As such, this study aimed to achieve two primary objectives: mapping the current geographical range of the Hyalomma vector tick, the key culprit in spreading the CCHF disease in Iraq; predicting its potential habitats; and outlining the environmental variables that play a significant role in determining the tick’s distribution. This study conducted an analysis using modeling techniques to identify key factors affecting the habitat suitability and distribution of the Hyalomma tick. The significant contributors to Hyalomma tick habitat preference were land cover (50.8%), elevation (30.4%), NDVI (5.7%), temperature seasonality (4.7%), precipitation seasonality (3.3%), sheep density (2.3%), and goat density (2.2%). Notably, topographic variables, especially land cover type (herbaceous wetland and spare vegetation) and elevation, played a crucial role in influencing the tick’s distribution. Land cover types, like spare vegetation and herbaceous wetland, were important habitats for Hyalomma ticks, providing suitable environments for their growth and development. These results are in line with previous research highlighting the importance of these factors [8,9]. These factors are related to other covariates, such as temperature, NDVI, and, hence, herd densities. NDVI plays a crucial role in indicating the presence of soil moisture that immature tick stages rely on [10]. In addition, NDVI is related to vegetation greenness and, hence, the attraction of livestock, such as sheep and goats. In support of this argument, the results of this study showed that herbaceous vegetation was highly suitable for the vector ticks in Iraq. Livestock density, particularly sheep and goat density, also influenced Hyalomma vector tick distribution, suggesting the ecoparaste nature of the ticks. The CCHF virus, Nairovirus genus, in the infected ticks undergoes multiplication and triggers viremia in sheep, cattle, and goats for a short period during which they do not exhibit any symptomatic illness [18].
Ticks are blood-feeding parasites that need one or more hosts to go through their life cycle. They spend a significant portion of their lives away from hosts, and during this time, environmental factors have a significant impact on their population dynamics [45]. The climate is one of the major factors affecting ticks, particularly the microclimate, which directly affects ticks more than the macroclimate. Researchers have often utilized macroclimatic temperature and rainfall data to create models predicting tick distribution [10,46,47] because microclimate data, particularly at large scales, are scarce or nonexistent. Ticks tend to seek out favorable levels of relative humidity in their immediate environments [48], where they might actively absorb water [49]. As cold-blooded organisms, temperature also influences their growth rates and activity [50]. Consequently, both temperature and water availability affect where ticks are found.

4.3. Spatial Distribution of Hyalomma Vector Ticks

In Iraq, only a fraction of the total land, approximately 51% of the region (225,665 km2 out of 441,724 km2), provides a suitable habitat for the species. Among these suitable areas, the majority, about 41.57% (183,631 km2), were deemed to have low suitability. Moderately suitable areas accounted for 8.61% (38,039 km2), while highly suitable areas were limited to just 0.9% (3994 km2). Regions posing a high risk for the spread of CCHF include the southern parts of the Dhi-Qar, Mysan, and Basra provinces. Additionally, some parts of Baghdad, the capital city, were also identified as risky areas for the dispersal of the vector tick, as demonstrated by the modeling (Figure 4). Moderately risky regions encompass certain sites across Diyala, Wasit, and Almuthanna in the southern part of Iraq, along with small areas in Duhok, Erbil, Kirkuk, and Ninaw in the north (Figure 4). Notably, significant portions of the country, particularly the north, northeast, and western areas of Anbar, were found to have unsuitable habitats for Hyalomma vector ticks in Iraq. Overall, the southern provinces, in comparison to the northeastern provinces, seem more suitable to the hard ticks to complete their life cycles. Hyalomma ticks are known to exhibit ditropic behavior. This means that after engorged larvae feed on a host, they stay on the same host to moult and later feed again as nymphs. As adults, they seek out and feed on a different host individual after a diapause period. The tick’s activity is generally highest during the summer months across all stages of its life cycle [2]. This could be one of the main reasons for the high number of CCHF case reports in Iraq. On the other hand, the southern provinces seem suitable for the ticks to successfully complete their life cycles on multiple hosts. This intricate behavior of relying on several hosts across various life stages presents difficulties in precisely predicting the future spread of these vectors.

4.4. Implications for Precautionary Measures

The main puzzle in biogeography and ecology involves understanding how various factors influence the distribution of a biological agent in a particular area. To tackle this challenge, species distribution models are often used. This method helps establish the link between the agent and its ecological needs, while also pinpointing crucial environmental factors that determine its presence in a specific geographic zone. By combining GIS-based spatial techniques and modeling approaches, this study provided valuable information about the current and potential habitat distribution of Hyalomma vector ticks. This information can aid management efforts and help decision-makers respond appropriately to prevent CCHF disease. This study created a detailed categorical map that highlights the potential distribution of Hyalomma ticks in Iraq. This map is crucial for public health, as it pinpoints areas that could be susceptible to outbreaks of CCHF disease, especially if the virus has expanded to new geographical areas. This research also predicted the probability of tick vectors appearing in regions that are conducive to the emergence of the CCHF. These results have immense value in implementing efficient disease surveillance and devising treatment strategies involving antiviral drugs and vector control programs, particularly in areas where regular national surveillance is absent.

5. Conclusions

The modeling results demonstrated that the southern provinces of Iraq, such as the Dhi-Qar, Mysan, and Basra provinces, are the most suitable areas for survival and, hence, the spread of the CCHF disease to other areas. Particular attention by the Ministry of Health and Environment and other related stakeholders should be given to herbaceous wetlands in the southern part of Iraq, specifically in the provinces of Dhi-Qar, Mysan, and Basra. The categorized map classes offer a meaningful tool for taking precautionary measures and controlling and minimizing the impacts of the disease. Land cover types, particularly herbaceous wetland, spare vegetation, and elevation, are the key factors for thriving vector ticks. Preventive measures should focus on the highly suitable class category, which is nearly 0.9% (3994 km2) of the total area of the country. This study is the first attempt in Iraq to model the spatial distribution of an endemic disease. Overall, this research contributes to the understanding of the Hyalomma vector tick habitat in an understudied region and emphasizes the importance of proactive preventive strategies to safeguard the spread of CCHF disease. This research offers novel foundational data regarding the current and potential geographical distribution of the Hyalomma vector tick in Iraq. Future studies should consider including microclimate data and individual identification of the tick species.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Not applicable.

Acknowledgments

The support of the Department of Biology and the College of Education at the University of Sulaimani is highly appreciated.

Conflicts of Interest

The author declares no conflict of interest.

References

  1. Alhilfi, R.A.; Khaleel, H.A.; Raheem, B.M.; Mahdi, S.G.; Tabche, C.; Rawaf, S. Large outbreak of Crimean-Congo haemorrhagic fever in Iraq, 2022. IJID Reg. 2023, 6, 76–79. [Google Scholar] [CrossRef] [PubMed]
  2. Hoogstraal, H. The epidemiology of tick-borne Crimean-Congo hemorrhagic fever in Asia, Europe, and Africa. J. Med. Entomol. 1979, 15, 307–417. [Google Scholar] [CrossRef] [PubMed]
  3. Al-Rubaye, D.; Al-Rubaye, T.S.; Shaker, M.; Naif, H.M. Recent outbreaks of crimean–congo hemorrhagic fever (CCHF) In Iraq. Sci. Arch. 2022, 3, 109–112. [Google Scholar] [CrossRef]
  4. Bente, D.A.; Forrester, N.L.; Watts, D.M.; McAuley, A.J.; Whitehouse, C.A.; Bray, M. Crimean-Congo hemorrhagic fever: History, epidemiology, pathogenesis, clinical syndrome and genetic diversity. Antivir. Res. 2013, 100, 159–189. [Google Scholar] [CrossRef]
  5. Alavi-Naini, R.; Moghtaderi, A.; Koohpayeh, H.-R.; Sharifi-Mood, B.; Naderi, M.; Metanat, M.; Izadi, M. Crimean-Congo hemorrhagic fever in Southeast of Iran. J. Infect. 2006, 52, 378–382. [Google Scholar] [CrossRef]
  6. ECDC. Crimean–Congo Haemorrhagic Fever—Annual Epidemiological Report for 2016. Available online: https://www.ecdc.europa.eu/sites/default/files/documents/crimean%E2%80%93congo-haemorrhagic-fever-annual-epidemiological-report-2016.pdf (accessed on 20 April 2023).
  7. Mild, M.; Simon, M.; Albert, J.; Mirazimi, A. Towards an understanding of the migration of Crimean–Congo hemorrhagic fever virus. J. Gen. Virol. 2010, 91, 199–207. [Google Scholar] [CrossRef]
  8. Deka, M.A. The geographic distribution of Crimean-Congo hemorrhagic fever in Tajikistan and central Asia. Pap. Appl. Geogr. 2017, 3, 68–84. [Google Scholar] [CrossRef]
  9. Deka, M.A. Crimean-Congo hemorrhagic fever geographic and environmental risk assessment in the Balkan and anatolian Peninsulas. Pap. Appl. Geogr. 2018, 4, 46–71. [Google Scholar] [CrossRef]
  10. Okely, M.; Anan, R.; Gad-Allah, S.; Samy, A.M. Mapping the environmental suitability of etiological agent and tick vectors of Crimean-Congo hemorrhagic fever. Acta Trop. 2020, 203, 105319. [Google Scholar] [CrossRef]
  11. Estrada-Pena, A.; Zatansever, Z.; Gargili, A.; Aktas, M.; Uzun, R.; Ergonul, O.; Jongejan, F. Modeling the spatial distribution of Crimean-Congo hemorrhagic fever outbreaks in Turkey. Vector-Borne Zoonotic Dis. 2007, 7, 667–678. [Google Scholar] [CrossRef]
  12. Williams, H.W.; Cross, D.E.; Crump, H.L.; Drost, C.J.; Thomas, C.J. Climate suitability for European ticks: Assessing species distribution models against null models and projection under AR5 climate. Parasites Vectors 2015, 8, 440. [Google Scholar] [CrossRef] [PubMed]
  13. Pascoe, E.L.; Nava, S.; Labruna, M.B.; Paddock, C.D.; Levin, M.L.; Marcantonio, M.; Foley, J.E. Predicting the northward expansion of tropical lineage Rhipicephalus sanguineus sensu lato ticks in the United States and its implications for medical and veterinary health. PLoS ONE 2022, 17, e0271683. [Google Scholar] [CrossRef] [PubMed]
  14. Orusa, T.; Orusa, R.; Viani, A.; Carella, E.; Borgogno Mondino, E. Geomatics and EO data to support wildlife diseases assessment at landscape level: A pilot experience to map infectious keratoconjunctivitis in chamois and phenological trends in Aosta Valley (NW Italy). Remote Sens. 2020, 12, 3542. [Google Scholar] [CrossRef]
  15. Ippoliti, C.; Candeloro, L.; Gilbert, M.; Goffredo, M.; Mancini, G.; Curci, G.; Falasca, S.; Tora, S.; Di Lorenzo, A.; Quaglia, M. Defining ecological regions in Italy based on a multivariate clustering approach: A first step towards a targeted vector borne disease surveillance. PLoS ONE 2019, 14, e0219072. [Google Scholar] [CrossRef]
  16. Martin, V.; Chevalier, V.; Ceccato, P.; Anyamba, A.; De Simone, L.; Lubroth, J.; De La Rocque, S.; Domenech, J. The impact of climate change on the epidemiology and control of Rift Valley fever. Rev. Sci. Tech. 2008, 27, 413–426. [Google Scholar] [CrossRef]
  17. Jafar, U.; Usman, M.; Ehsan, M.; Naveed, A.; Ayyan, M.; Cheema, H.A. The outbreak of Crimean-Congo hemorrhagic fever in Iraq-Challenges and way forward. Ann. Med. Surg. 2022, 81, 104382. [Google Scholar] [CrossRef] [PubMed]
  18. Majeed, B.; Dicker, R.; Nawar, A.; Badri, S.; Noah, A.; Muslem, H. Morbidity and mortality of Crimean-Congo hemorrhagic fever in Iraq: Cases reported to the National Surveillance System, 1990–2010. Trans. R. Soc. Trop. Med. Hyg. 2012, 106, 480–483. [Google Scholar] [CrossRef] [PubMed]
  19. Tantawi, H.; Al-Moslih, M.; Al-Janabi, N.; Al-Bana, A.; Mahmud, M.; Jurji, F.; Yonan, M.; Al-Ani, F.; Al-Tikriti, S. Crimean-Congo haemorrhagic fever virus in Iraq: Isolation, identification and electron microscopy. Acta Virol. 1980, 24, 464–467. [Google Scholar]
  20. Tantawi, H.; Shony, M.; Al-Tikriti, S. Antibodies to Crimean-Congo haemorrhagic fever virus in domestic animals in Iraq: A seroepidemiological survey. Int. J. Zoonoses 1981, 8, 115–120. [Google Scholar]
  21. Malinowski, J.C. Iraq: A Geography; United States Military Academy: West Point, NY, USA, 2002. [Google Scholar]
  22. Hashim, B.M.; Al-Naseri, S.K.; Al Maliki, A.; Sa’adi, Z.; Malik, A.; Yaseen, Z.M. On the investigation of COVID-19 lockdown influence on air pollution concentration: Regional investigation over eighteen provinces in Iraq. Environ. Sci. Pollut. Res. 2021, 28, 50344–50362. [Google Scholar] [CrossRef]
  23. Brown, J.L.; Bennett, J.R.; French, C.M. SDMtoolbox 2.0: The next generation Python-based GIS toolkit for landscape genetic, biogeographic and species distribution model analyses. PeerJ 2017, 5, e4095. [Google Scholar] [CrossRef] [PubMed]
  24. Støa, B.; Halvorsen, R.; Stokland, J.N.; Gusarov, V.I. How much is enough? Influence of number of presence observations on the performance of species distribution models. Sommerfeltia 2019, 39, 1–28. [Google Scholar] [CrossRef]
  25. Wisz, M.S.; Hijmans, R.; Li, J.; Peterson, A.T.; Graham, C.; Guisan, A.; NCEAS Predicting Species Distributions Working Group. Effects of sample size on the performance of species distribution models. Divers. Distrib. 2008, 14, 763–773. [Google Scholar] [CrossRef]
  26. Hijmans, R.J.; Cameron, S.E.; Parra, J.L.; Jones, P.G.; Jarvis, A. Very high resolution interpolated climate surfaces for global land areas. Int. J. Climatol. A J. R. Meteorol. Soc. 2005, 25, 1965–1978. [Google Scholar] [CrossRef]
  27. Gilbert, M.; Cinardi, G.; Da Re, D.; Wint, W.G.R.; Wisser, D.; Robinson, T.P. Global Sheep Distribution in 2015 (5 min of Arc); Harvard Dataverse, V1; Harvard: Cambridge, MA, USA, 2022. [Google Scholar] [CrossRef]
  28. Buchhorn, M.; Lesiv, M.; Tsendbazar, N.-E.; Herold, M.; Bertels, L.; Smets, B. Copernicus global land cover layers—Collection 2. Remote Sens. 2020, 12, 1044. [Google Scholar] [CrossRef]
  29. Dormann, C.F.; Elith, J.; Bacher, S.; Buchmann, C.; Carl, G.; Carré, G.; Marquéz, J.R.G.; Gruber, B.; Lafourcade, B.; Leitão, P.J. Collinearity: A review of methods to deal with it and a simulation study evaluating their performance. Ecography 2013, 36, 27–46. [Google Scholar] [CrossRef]
  30. Elith, J.; Graham, C.H.; Anderson, R.P.; Dudík, M.; Ferrier, S.; Guisan, A.; Hijmans, R.J.; Huettmann, F.; Leathwick, J.R.; Lehmann, A.; et al. Novel methods improve prediction of species’ distributions from occurrence data. Ecography 2006, 29, 129–151. [Google Scholar] [CrossRef]
  31. Phillips, S.J.; Anderson, R.P.; Dudík, M.; Schapire, R.E.; Blair, M.E. Opening the black box: An open-source release of Maxent. Ecography 2017, 40, 887–893. [Google Scholar] [CrossRef]
  32. Khwarahm, N.R. Mapping current and potential future distributions of the oak tree (Quercus aegilops) in the Kurdistan Region, Iraq. Ecol. Process. 2020, 9, 1–16. [Google Scholar] [CrossRef]
  33. Phillips, S.J.; Dudík, M.; Schapire, R.E. A maximum entropy approach to species distribution modeling. In Proceedings of the Twenty-First International Conference on Machine Learning, Banff, AB, Canada, 4–8 July 2004; p. 83. [Google Scholar]
  34. Morales, N.S.; Fernández, I.C.; Baca-González, V. MaxEnt’s parameter configuration and small samples: Are we paying attention to recommendations? A systematic review. PeerJ 2017, 5, e3093. [Google Scholar] [CrossRef]
  35. Elith, J.; Leathwick, J.R. Species distribution models: Ecological explanation and prediction across space and time. Annu. Rev. Ecol. Evol. Syst. 2009, 40, 677–697. [Google Scholar] [CrossRef]
  36. Merow, C.; Smith, M.J.; Silander, J.A., Jr. A practical guide to MaxEnt for modeling species’ distributions: What it does, and why inputs and settings matter. Ecography 2013, 36, 1058–1069. [Google Scholar] [CrossRef]
  37. Radha, K.O.; Khwarahm, N.R. An Integrated Approach to Map the Impact of Climate Change on the Distributions of Crataegus azarolus and Crataegus monogyna in Kurdistan Region, Iraq. Sustainability 2022, 14, 14621. [Google Scholar] [CrossRef]
  38. Hama, A.A.; Khwarahm, N.R. Predictive mapping of two endemic oak tree species under climate change scenarios in a semiarid region: Range overlap and implications for conservation. Ecol. Inform. 2023, 73, 101930. [Google Scholar] [CrossRef]
  39. HamadAmin, B.A.; Khwarahm, N.R. Mapping Impacts of Climate Change on the Distributions of Two Endemic Tree Species under Socioeconomic Pathway Scenarios (SSP). Sustainability 2023, 15, 5469. [Google Scholar] [CrossRef]
  40. Chauvier, Y.; Thuiller, W.; Brun, P.; Lavergne, S.; Descombes, P.; Karger, D.N.; Renaud, J.; Zimmermann, N.E. Influence of climate, soil, and land cover on plant species distribution in the European Alps. Ecol. Monogr. 2021, 91, e01433. [Google Scholar] [CrossRef]
  41. Shabani, F.; Kumar, L.; Ahmadi, M. A comparison of absolute performance of different correlative and mechanistic species distribution models in an independent area. Ecol. Evol. 2016, 6, 5973–5986. [Google Scholar] [CrossRef]
  42. Thuiller, W.; Lafourcade, B.; Engler, R.; Araújo, M.B. BIOMOD–A platform for ensemble forecasting of species distributions. Ecography 2009, 32, 369–373. [Google Scholar] [CrossRef]
  43. Zhao, G.-P.; Wang, Y.-X.; Fan, Z.-W.; Ji, Y.; Liu, M.-j.; Zhang, W.-H.; Li, X.-L.; Zhou, S.-X.; Li, H.; Liang, S. Mapping ticks and tick-borne pathogens in China. Nat. Commun. 2021, 12, 1075. [Google Scholar] [CrossRef]
  44. Gray, J.; Dautel, H.; Estrada-Peña, A.; Kahl, O.; Lindgren, E. Effects of climate change on ticks and tick-borne diseases in Europe. Interdiscip. Perspect. Infect. Dis. 2009, 2009, 593232. [Google Scholar] [CrossRef]
  45. Estrada-Peña, A.; de la Fuente, J. The ecology of ticks and epidemiology of tick-borne viral diseases. Antivir. Res. 2014, 108, 104–128. [Google Scholar] [CrossRef] [PubMed]
  46. Estrada-Pena, A.; Venzal, J.M. Climate niches of tick species in the Mediterranean region: Modeling of occurrence data, distributional constraints, and impact of climate change. J. Med. Entomol. 2007, 44, 1130–1138. [Google Scholar] [CrossRef] [PubMed]
  47. Porretta, D.; Mastrantonio, V.; Amendolia, S.; Gaiarsa, S.; Epis, S.; Genchi, C.; Bandi, C.; Otranto, D.; Urbanelli, S. Effects of global changes on the climatic niche of the tick Ixodes ricinus inferred by species distribution modelling. Parasites Vectors 2013, 6, 271. [Google Scholar] [CrossRef] [PubMed]
  48. Ostfeld, R.S.; Brunner, J.L. Climate change and Ixodes tick-borne diseases of humans. Philos. Trans. R. Soc. B Biol. Sci. 2015, 370, 20140051. [Google Scholar] [CrossRef]
  49. Perret, J.-L.; Guigoz, E.; Rais, O.; Gern, L. Influence of saturation deficit and temperature on Ixodes ricinus tick questing activity in a Lyme borreliosis-endemic area (Switzerland). Parasitol. Res. 2000, 86, 554–557. [Google Scholar] [CrossRef]
  50. Randolph, S. Predicting the risk of tick-borne diseases. Int. J. Med. Microbiol. 2002, 291, 6–10. [Google Scholar] [CrossRef]
Sustainability 15 13669 g001
Figure 1. Location of the study site and the extent of the regional boundaries.
Figure 1. Location of the study site and the extent of the regional boundaries.
Sustainability 15 13669 g001
Sustainability 15 13669 g002
Figure 2. Area under the Receiver Operating Characteristic (ROC) curve for the training data, averaged over the replicate runs. The specificity is based on the predicted area. The average training AUC for the replicate runs is 0.885, with a standard deviation of 0.026.
Figure 2. Area under the Receiver Operating Characteristic (ROC) curve for the training data, averaged over the replicate runs. The specificity is based on the predicted area. The average training AUC for the replicate runs is 0.885, with a standard deviation of 0.026.
Sustainability 15 13669 g002
Sustainability 15 13669 g003aSustainability 15 13669 g003b
Figure 3. The gains achieved through regularized training (in percentage), (top figure), and AUC (Area Under the Curve) gains (in percentage), (bottom figure), were analyzed by the Jackknife test to understand the relative impact of the variables on the suitability distribution of Hyalomma vector ticks in Iraq.
Figure 3. The gains achieved through regularized training (in percentage), (top figure), and AUC (Area Under the Curve) gains (in percentage), (bottom figure), were analyzed by the Jackknife test to understand the relative impact of the variables on the suitability distribution of Hyalomma vector ticks in Iraq.
Sustainability 15 13669 g003aSustainability 15 13669 g003b
Sustainability 15 13669 g004
Figure 4. Model output depicting the categorized spatial distribution of Hyalomma ticks in Iraq.
Figure 4. Model output depicting the categorized spatial distribution of Hyalomma ticks in Iraq.
Sustainability 15 13669 g004
Table 1. Matrix showing the correlations among the variables utilized in the modeling process.
Table 1. Matrix showing the correlations among the variables utilized in the modeling process.
12345678
11.00−0.04−0.36−0.38−0.29−0.44−0.460.13
2−0.041.000.240.11−0.37−0.050.31−0.50
3−0.360.241.000.280.330.450.69−0.34
4−0.380.110.281.000.280.340.43−0.15
5−0.29−0.370.330.281.000.390.46−0.10
6−0.44−0.050.450.340.391.000.60−0.19
7−0.460.310.690.430.460.601.00−0.45
80.13−0.50−0.34−0.15−0.10−0.19−0.451.00
1 = Land cover 2019, 2 = DEM, 3 = NDVI, 4 = Temperature seasonality, 5 = Precipitation seasonality, 6 = Sheep density, 7 = Goat density, 8 = Mean diurnal range (temperature).
Table 2. The factors utilized in constructing the model, along with their abbreviated codes, units, percentage of contribution (the percentage of contribution of a variable that influences the likelihood of Hyalomma vector tick distribution), and permutation importance. Permutation importance highlights how the values of variables, which are randomly shuffled in the training data of both presence occurrences and background points, contribute subsequently.
Table 2. The factors utilized in constructing the model, along with their abbreviated codes, units, percentage of contribution (the percentage of contribution of a variable that influences the likelihood of Hyalomma vector tick distribution), and permutation importance. Permutation importance highlights how the values of variables, which are randomly shuffled in the training data of both presence occurrences and background points, contribute subsequently.
VariableCode/UnitPercent ContributionPermutation Importance
Land cover for 2019landcover201950.817.8
Digital Elevation Model DEM/(meter)30.449.5
Normalized Difference Vegetation IndexIRQ_ndvi5.79.4
Temperature seasonality wc2.0_bio_30s_04/(°C)4.72.7
Precipitation seasonality (coefficient of variation)wc2.0_bio_30s_15/(mm)3.310.6
Sheep density c5_sh_2015_da/(sheep/area (km)2.37.7
Goat density c5_gt_2015_da/(goat/area (km)2.21
Mean diurnal range (max temp–min temp)wc2.0_bio_30s_02/(°C)0.51.3
Table 3. Habitat suitability categorization of Hyalomma vector ticks in Iraq; the suitable regions were categorized into low, medium, and high suitability, each measured in square kilometers along with their corresponding percentages.
Table 3. Habitat suitability categorization of Hyalomma vector ticks in Iraq; the suitable regions were categorized into low, medium, and high suitability, each measured in square kilometers along with their corresponding percentages.
Suitability ClassArea (km2)Area (%)
Unsuitable216,059.6048.91
Low suitable183,631.7541.57
Medium suitable38,039.058.61
High suitable3994.370.90
Total441,724.78100.00
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Khwarahm, N.R. Predicting the Spatial Distribution of Hyalomma ssp., Vector Ticks of Crimean–Congo Haemorrhagic Fever in Iraq. Sustainability 2023, 15, 13669. https://doi.org/10.3390/su151813669

AMA Style

Khwarahm NR. Predicting the Spatial Distribution of Hyalomma ssp., Vector Ticks of Crimean–Congo Haemorrhagic Fever in Iraq. Sustainability. 2023; 15(18):13669. https://doi.org/10.3390/su151813669

Chicago/Turabian Style

Khwarahm, Nabaz R. 2023. "Predicting the Spatial Distribution of Hyalomma ssp., Vector Ticks of Crimean–Congo Haemorrhagic Fever in Iraq" Sustainability 15, no. 18: 13669. https://doi.org/10.3390/su151813669

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop