Larval Anopheles Species Composition and Diversity at Different Habitats and Seasons of Gondar Zuria District, Ethiopia

Anopheles species' larval habitats are diversified and season dependent. Anopheles larvae can be found at different habitats and their preference may vary seasonally. Knowledge of species diversity and distribution helps plan malaria control interventions. Anopheles larvae were sampled using the WHO standard 350 ml dipper from breeding habitats of irrigation, pond, sewage, stream, and swamp. The collected sample larvae were identified microscopically to species using morphological keys. Among Anopheles species recorded, Anopheles gambiae complex was the most abundant followed by Anopheles christyi, Anopheles cinereus, Anopheles demeilloni, and Anopheles pharoensis in descending order. Anopheles species occurred more in January than in other months of the study period and less in March and April. For any particular mosquito species, larval abundance did not significantly vary between the habitats; in other words, all habitats contributed equally. In this study, we confirmed that Anopheles mosquito larval population varied more with respect to species than to habitats and months. Interventions could be launched targeting each habitat; during the month, numbers were high.


Introduction
Female Anopheles mosquitoes transmit deadly malaria disease-causing pathogens [1,2]. To date, globally, about 500 Anopheles species are listed. Among the hundreds of identified Anopheles species, 30-40 species are responsible for malaria transmission, of which about 15 species are the major ones [3][4][5]. Anopheles gambiae complex, An. funestus, An. nili, An. pharoensis, and An. moucheti are malaria vectors in Africa [6]. An. gambiae complex and An. funestus complexes are the dominant malaria vector species widely distributed in Africa [4,6]. An. gambiae complex accounts for the majority of malaria cases and deaths [7]. An. arabiensis is the major malaria vector widely distributed in Ethiopia [8][9][10][11]. Anopheles species have diversified larval habitats; in some cases, different species share the same habitats, while other species are more selective for habitats with specific characteristics [12][13][14][15]. Spatial and temporal distributions of Anopheles species larvae are affected by habitat types and environmental factors including temperature, humidity, and precipitation across seasons [16][17][18][19]. Studies reported associations between habitat types of different Anopheles species and their abundance varied across months [14,16,[19][20][21]. However, in the current study area, there is no information on the species diversity and abundance of Anopheles mosquito larvae in relation to habitat types and seasons. e objective of this study was to identify Anopheles species, determine their association to different habitats-swamp, pond, stream, sewage, and irrigation canal-as well as determine seasonal variation.

Materials and Methods
Anopheles larval sampling was made in potential breeding habitats at Gondar Zuria District (37°24′ 24″E-37°45′ 43″ E and 12°7′ 23″ N-12°39′24″N), 1500-2300 m above sea level ( Figure 1). e district experiences mild temperatures from 11 to 32°C. Mean annual rainfall ranges between 900 mm and 1035 mm, which is bimodal, with short rains from March to May, long rains from June to September, post rains from October to November, and dry season from December to February [22]. Larvae of Anopheles species were collected using the WHO standard 350 ml dipper from breeding habitats of irrigation, pond, sewage, stream, and swamp. Sampling was conducted in the dry season (December/ 2019-February 2020) and the short rainy season (March--May/2020). Ten technical replicates (dips) on 10 subsites of each habitat type were made to collect Anopheles larvae monthly during the study period [16]. e collected larvae were preserved in 70% alcohol in separate coded vials within 24 h of collection. Species were identified using dichotomous keys and recorded with respect to habitat and season [23,24].

Data Analyses.
Data were subjected to analysis of variance using SAS software (version 9.4) for determining differences in larval abundance between habitats, between months, and their interaction, where appropriate. Detrended correspondence analysis was performed to determine the association between species and habitats and between species and sampling months. And finally, a number of species diversity indices were computed to establish the diversity of species at different habitats and months. Correspondence analysis and diversity indices were computed using PAST 4.03 software [25]. e following 11 diversity indices were used in the present investigation. (7) Fisher's alpha is a diversity index, defined implicitly by the formula S � α ln(1 + n/α), where S is the number of taxa, n is the number of individuals, and α is Fisher's alpha. Fisher's α measures diversity within a population, and it is considered the best diversity index for many communities of species, including Lepidoptera.  According to the results of the analysis of variance, while the interaction effect of habitat and larval species was significant (F � 33.03, DF � 24,125, P < 0.0001), it was not the case between months and species of Anopheles mosquito numbers (F � 0.98, DF � 16, P � 0.4802).

Distribution of Larvae at Different
None of the mosquito species significantly varied in abundance between different habitats (Table 1). Regardless of habitat, An. gambiae complex had the highest larval abundance, followed by An. christyi, An. cinereus, An. demeilloni, and An. pharoensis, in order of importance.

Seasonal Variation of Anopheles Larvae.
According to the result of the analysis of variance, the number of larvae varied significantly with respect to months (F � 33.3, DF � 29,120, P < 0.0001). Numbers were slightly higher in January and lower in March and April ( Table 2).
Regardless of differences in sampling months, An. gambiae complex had the highest larval abundance, followed by An. christyi, An. cinereus, An. demeilloni, and An. pharoensis, in order of importance.

Detrended Correspondence between Mosquito Larval
Species and Habitats. In the case of habitats, the correspondence analysis explained 100% of the association between habitats and species, with 80% in dimension 1 and 20% in dimension 2 (Figure 2, top). Dimension 1 was a result of the opposite association between irrigation and sewage and dimension 2 was that of irrigation and pond or puddle and swamp. As to the species, most of them were found close to the centroid (origin) which means that they all had similar trends, although there was a tendency to align themselves on different sides of dimension 1, showing some differences between them. None of the species appeared to be associated with any of the habitats.

Detrended Correspondence between Mosquito Larval Species and Months.
e correspondence analysis generally accounted for 100% of the association between species and the habitats (Figure 2, bottom). Dimension 1 accounted for 99.9% of the association between time (month) and a number of individuals of each Anopheles species corresponding to January and February (left) vs. May and December (left). Dimension 2 explained nothing for the association.
Most of the Anopheles species appeared to be spaced not very far from the origin, indicating a similar contribution and none of them had an affinity to any month.

Diversity Indices.
According to the results of the Shannon-Weiner index, the species diversity of Anopheles mosquito larvae appeared to be similar across the different habitats (around H′ � 1.5).
is happened  Journal of Tropical Medicine because all habitats had the same number of species, i.e., 5. ere was no dominance of any species in any habitat during this study (D � 0.23). A similar condition was found for Simpson_1_D across the different habitats. Equitability (J) was high indicating that species were equitably represented in all habitats (J � 0.94). Other indices did not vary across the different habitats (Table 3).
When the same thing was seen for sampling date across the season, a similar pattern was observed as the one seen on habitats, with negligible differences (Table 4).

Discussion
In this study, we confirmed that the different habitats had similar abundance of Anopheles mosquito larvae, while the different mosquito species varied in abundance significantly, for example, An. gambiae complex was the most abundant of all. e field collection of Anopheles larvae in the current study indicated the presence of Anopheles species including An. gambiae complex, presumably An. arabiensis, An. christyi, An. cinereus, An. demeilloni, and An. pharoensis.  Among the mosquito species recorded, An. gambiae complex was the dominant species and it had a stronger affinity to the stream habitat than the other habitats. According to the detrended correspondence analysis, none of the mosquito species in the study area were found close to any of the habitats. erefore, the present study indicated that all natural habitats contributed almost equally and no habitat was particularly associated with any of the mosquito species.
In line with the present study, various workers reported that An. gambiae complex was the most abundant species in different parts of Africa and Ethiopia [4,6,7,[26][27][28][29]. e better occurrence of the Anopheles species in January could be because the low temperature of the study area is not optimum for frequent adult emergence and the larvae stay a long time in breeding habitats [17][18][19][20]30].
Similar to the present study, the larvae of An. gambiae complex prefers small sun-exposed habitats mostly associated with slow-moving streams due to the comparative stability of the habitats with the optimum temperature where the female mosquito prefers to lay her egg and the eggs hatch in these conditions [19,20,27,31,32].
In contrast, Hinne et al. [29] reported that ponds of dugout wells were the most productive habitats of Anopheles mosquitoes during the dry season, while swamps were also productive both in dry and rainy seasons in Africa. On the other hand, Getachew et al. [27] reported no association between the primary malaria mosquito, An. gambiae complex with swamps. Tire tracks, puddles, and collected rainwater supported the breeding of Anopheles mosquitoes temporarily [27].
On the other hand, irrigation-associated habitats were the major sources of Anopheles mosquitoes such as An. arabiensis and An. pharoensis [33,34]. According to previous reports, An. christyi, An. cinereus, and An. demeilloni were mostly found in streams and swamps, whereas An. pharoensis were associated with swamp habitats [32].
In the current study, Anopheles species diversity was found to be average, i.e., H′ � 1.5, indicating some signs of environmental degradation that failed to support more diversity. In another study carried out in Mara River and its tributaries, Kenya and Tanzania, Dida et al. [19] reported lower diversity than the current one. As a result, in the current study, no species were found particularly dominant (dominance�0.24 overall), which is also supported by the high evenness and equitability values (>0.9) indicating the lack of any dominant species.
In the current study, it was possible to underpin that it was not the habitats that mattered but the species themselves that dictated mosquito larval abundance because the larval abundance of any mosquito species did not vary between habitats. Significant differences in abundance were observed among the different species because of their intrinsic differences in their bionomic potential. Abundance slightly varied between months, whereby numbers were more in January than in March and April. Further study is required  to find out the year-round species composition of Anopheles mosquitoes and molecular identification of sibling species of An. gambiae complex in the area. Moreover, further study is required to determine the insecticide resistance level of the primary malaria mosquito, An. gambiae complex at a broader geographic area.

Conclusion
e present study gave information about Anopheles species diversity at different habitats during the different months.
e results indicate that permanent and semipermanent larval habitats exist for malaria transmission, where vector management tools could be applied. Seasonal variation was noticed as larval population declined in March and April. Anopheles species diversity was rather low (about H′ � 1.5) as well as dominance (about 0.24 across all habitats and months). As a result, evenness and equitability were high (above 0.9) indicating the lack of any dominant species. e detrended correspondence analysis showed associations between habitats and or months, but none of the mosquito species had an affinity for any habitat or month.

Data Availability
e data used to support this study are available from the corresponding author upon request.

Consent
All parties including the authors agree on its publication in the current form. Disclosure e funder had no role in this study.

Conflicts of Interest
e authors declare that there are no conflicts of interest.

Authors' Contributions
YA initiated the research agenda, supervised the study, and prepared a draft manuscript. SA conducted the field and laboratory works, collected data, and wrote the draft thesis. MW conducted the statistical analyses and critically revised the final manuscript. All authors read and approved the final version of the manuscript.