Larval habitats diversity and productivity
A total of 415 mosquito breeding habitats were identified both inside (317) and outside (98) houses. Out of these, 168 (40.5%) were found in Kilifi, 114 (27.5%) in Malindi and 133 (32.04%) in Mombasa (Table 1). Fourteen different larval habitats were identified and sampled in the three study sites. Overall, the most prevalent breeding habitats in the three sites were Jericans (66.9%), followed by water tanks (10.6%), small containers (6.75%) and drainage channels (6.02%). Other habitats encountered but in small numbers were: buckets, basins, ditches, water troughs, flower pots, swimming pools, chambers and earthen water pots (Table 1). In Kilifi, seven different types of mosquito-breeding habitats were identified. The most abundant habitat types were jericans (61.90%), followed by water tank (16.07%), assorted small containers (11.31%), drainages (6.55%), and tyres (1.79%). Others that were reported in small numbers include ditches and drainage chambers. Similarly, in Malindi, eleven different habitat types were identified (Table 1). The most abundant habitat type was jericans (71.05%) followed by small containers (7.02%), water tanks (5.26%), drainages (5.26%) and tyres (0.88%). Other habitats encountered though in small numbers include bucket, basin, ditch water trough, flowerpot and swimming pool (Table 1). In Mombasa, ten different habitats were surveyed and identified; jericans (69.92%) were the most abundant habitat type followed by water tanks (8.27%), drainages (6.02%), drums (5.26),
Table 1
Summary of the habitat productivity for indoor and outdoor locations in the three sites of urban coastal Kenya
| Kilifi | Malindi | Mombasa |
| Indoor | Outdoor | Indoor | Outdoor | Indoor | Outdoor |
Habitat Type | No of Habitats (+ ve habitats) | Mean Larvae | Mean Pupae | No of Habitats (+ ve habitats) | Mean Larvae | Mean Pupae | No of Habitats (+ ve habitats) | Mean Larvae | Mean Pupae | No of Habitats (+ ve habitats) | Mean Larvae | Mean Pupae | No of Habitats (+ ve habitats) | Mean Larvae | Mean Pupae | No of Habitats (+ ve habitats) | Mean Larvae | Mean Pupae |
Small container | 9 (1) | 10 | 0 | 10 (5) | 108 | 46 | 0 (0) | 0 | 0 | 8 (2) | 27 | 0 | 0 (0) | 0 | 0 | 1 (0) | 0 | 0 |
Drum | 0 (0) | 0 | 0 | 0 (0) | 0 | 0 | 0 (0) | 0 | 0 | 0 (0) | 0 | 0 | 7 (6) | 43 | 3 | 0 (0) | 0 | 0 |
Water tank | 24 (6) | 9 | 1 | 3 (1) | 15 | 0 | 3 (0) | 0 | 0 | 3 (2) | 74 | 70 | 9 (3) | 44 | 5 | 2 (1) | 41 | 0 |
Jerican | 104 (2) | 7 | 1 | 0 (0) | 0 | 0 | 62 (1) | 26 | 4 | 19 (1) | 50 | 0 | 91 (6) | 28 | 1 | 2 (0) | 0 | 0 |
Bucket | 0 (0) | 0 | 0 | 0 (0) | 0 | 0 | 0 (0) | 0 | 0 | 3 (2) | 58 | 50 | 2 (0) | 0 | 0 | 0 (0) | 0 | 0 |
Basin | 0 (0) | 0 | 0 | 0 (0) | 0 | 0 | 0 (0) | 0 | 0 | 1 (0) | 0 | 0 | 3 (0) | 0 | 0 | 0 (0) | 0 | 0 |
Drainage channels | 2 (0) | 0 | 0 | 9 (3) | 87 | 4 | 0 (0) | 0 | 0 | 6 (0) | 0 | 0 | 0 (0) | 0 | 0 | 8 (1) | 17 | 4 |
Ditch | 0 (0) | 0 | 0 | 1 (0) | 0 | 0 | 0 (0) | 0 | 0 | 1 (0) | 0 | 0 | 0 (0) | 0 | 0 | 2 (0) | 0 | 0 |
Tyre | 0 (0) | 0 | 0 | 3 (0) | 0 | 0 | 0 (0) | 0 | 0 | 1 (1) | 20 | 0 | 0 (0) | 0 | 0 | 5 (1) | 30 | 0 |
Water trough | 0 (0) | 0 | 0 | 0 (0) | 0 | 0 | 0 (0) | 0 | 0 | 3 (3) | 36 | 0 | 0 (0) | 0 | 0 | 0 (0) | 0 | 0 |
Flower pot | 0 (0) | 0 | 0 | 0 (0) | 0 | 0 | 0 (0) | 0 | 0 | 3 (0) | 0 | 0 | 0 (0) | 0 | 0 | 0 (0) | 0 | 0 |
Swimming pool | 0 (0) | 0 | 0 | 0 (0) | 0 | 0 | 0 (0) | 0 | 0 | 1 (0) | 0 | 0 | 0 (0) | 0 | 0 | 0 (0) | 0 | 0 |
Chambers | 0 (0) | 0 | 0 | 3 (0) | 0 | 0 | 0 (0) | 0 | 0 | 0 (0) | 0 | 0 | 0 (0) | 0 | 0 | 0 (0) | 0 | 0 |
Water pot | 0 (0) | 0 | 0 | 0 (0) | 0 | 0 | 0 (0) | 0 | 0 | 0 (0) | 0 | 0 | 1 (0) | 0 | 0 | 0 (0) | 0 | 0 |
Total | 139 (9) | 26 | 2 | 29 (9) | 210 | 50 | 65 (1) | 26 | 4 | 49 (10) | 216 | 120 | 113 (15) | 115 | 9 | 20 (3) | 88 | 4 |
and tyres (3.76%). Other habitat types that were encountered though in smaller quantities include: assorted small containers, bucket, basin, ditch and water pot (Table 1).
Overall, the most productive habitats indoors were drums, small containers, jericans and water tanks whereas for outdoors the most productive containers were drainage channels, small containers, tyres, water tanks, jericans and water troughs (Fig. 2). There was a significant association between habitat type and immature productivity (= p < 0.001). Productivity in this case can be defined as the efficiency of a habitat/container to produce larvae. A total of 18 breeding habitats in Kilifi, (6% indoors, 31%outdoors) were positive for mosquito immature stages. A T-test showed that there was significant difference in the density of immatures between indoor and outdoor (P < 0.05). The most productive indoor habitats in Kilifi were small containers, water tanks and jericans whereas outdoors were drainage channels, small containers and water tanks (Table 1). In Malindi, 12 habitats (2% indoors, 22% outdoors) were found to be positive for mosquito immatures and a T- test statistic indicated that there was no significant difference between indoor and outdoor positive habitats (P > 0.05). The most productive indoor habitats in Malindi were only jericans. On the other hand, the most productive habitats outdoors were water tanks, jericans, small containers and the least were tyres (Table 1). In Mombasa, 18 habitats (13% indoors, 15% outdoors) were found to be positive for mosquito immatures (Table 1). there was no significant difference in the density of immatures between indoor and outdoor (P > 0.05). In Mombasa, the most productive indoor habitats were water tanks, drums and jericans whereas for outdoor habitats, water tank was the most productive habitat type, followed by tyres and the least were drainage channels (Table 1).
Species composition in larval habitat collection
Overall, 889 adult mosquitoes belonging to two genera (Aedes and Culex) emerged from the larval population collected. Majority were Aedes aegypti (85.3%) and the rest being Culex quinquefasciatus (12.60%), Ae. vittatus (1.12%), and Cx. zombiensis (1.01%). Indoor immatures resulted in purely and exclusively Ae. aegypti mosquitoes whereas as outdoor had both Ae. aegypti and Cx. quinquefasciatus mosquitoes.
Larval infestation indices
Fifty-five (55) houses were sampled from the three sites for mosquito habitats. Out of these houses, 18 had containers that were positive for Ae. aegypti immatures, giving an overall House Index (HI)of 32.72%. A total of 317 containers were inspected indoors giving an overall Container Index (CI) of 8% and Breteau Index (BI) of 45.45. Mombasa had the highest indices (HI of 71.43, CI of 13.27 and BI of 107.14) compared to Malindi and Kilifi (Table 2).
Table 2
Indoor site-specific House, Container and Breteau indices for the 3 coastal urban area
Sampled site | No. houses of sampled houses | No. of positive houses | HI | No. of wet habitats | No. of positive habitats | CI | BI |
Kilifi | 30 | 7 | 23.33 | 139 | 9 | 6.47 | 30.00 |
Malindi | 11 | 1 | 9.09 | 65 | 1 | 1.54 | 9.09 |
Mombasa | 14 | 10 | 71.43 | 113 | 15 | 13.27 | 107.14 |
Overall | 55 | 18 | 32.73 | 317 | 25 | 7.89 | 45.45 |
Mosquito egg survivorship in dry habitats
A total of 29 dry habitat substrate/ dry habitat soil samples were collected from water tank (n = 2), small container (n = 1), tyres (n = 16) and flower pots (n = 10). Overall, 62% (n = 18) of the soil samples collected from the two sites (Kilifi and Malindi) were positive for larvae. Five hundred and six (506) adult mosquitoes resulted from the larvae reared from the dry breeding habitats substrate. Three Aedes species were identified including Aedes aegypti (98.4%), Aedes hirsutus (1.4%) and Aedes vittatus (0.2%) (Table 3).
Table 3
Summary of the soil samples collected from different container, positive habitats and the mosquito species that emerged
Site | Habitat type | No. of habitats ( % +ve) | Mosquito species | Total adults Emerged |
Kilifi | Tyre | 7 (57) | Ae. aegypti | 319 |
Ae. hirsutus | 7 |
Ae. vittatus | 1 |
Flower pot | 3 (66) | Ae. aegypti | 2 |
Water tank | 2 (0) | - | 0 |
Malindi | Small container | 1 (100) | Ae. aegypti | 16 |
Tyre | 9 (77) | Ae. aegypti | 79 |
flower pot | 7 (71) | Ae. aegypti | 82 |
Adult mosquito distribution and abundance collections
The relative abundance of adult mosquitoes collected indoors and outdoors by the Biogents Sentinel (BG) traps and Light traps (LT) is summarized in Table 4. Overall, 3,264 mosquitoes belonging to three genera (Culicines, Aedes and Anopheles) and 10 species were collected. Ae aegypti (838) and Cx. quinquefasciatus (2,364) were the most common species, and the least were Ae. mcintoshi, Ae. pembaensis, and Cx. annulioris, (n = 1). Cx. quinquefasciatus were mostly collected indoors (n = 2,140) compared to outdoors (n = 260) while more Ae. aegypti mosquitoes were captured outdoors (n = 816) compared to indoors (n = 22) (Table 4). Shannon diversity index (H) and evenness (EH) of mosquito species indicated a higher species diversity in Kilifi (H = 0.840) compared to Malindi (H = 0.662) and Mombasa (H = 0.385). Mosquitoes were evenly distributed in Kilifi (EH = 0.469) compared to Malindi (EH = 0.370) and Mombasa (EH = 0.215).
Table 4
Relative abundance of mosquito species collected using light trap (indoor) and BG sentinel trap (outdoor) in the three urban areas of coastal Kenya
| Light trap (indoor) | BG sentinel trap (outdoor) | |
Mosquito species | Kilifi | Malindi | Mombasa | Kilifi | Malindi | Mombasa | Total |
Ae. aegypti | 3 | 9 | 10 | 306 | 446 | 64 | 838 |
Ae. hirsutus | 3 | 0 | 0 | 0 | 0 | 0 | 3 |
Ae. mcintoshi | 1 | 0 | 0 | 0 | 0 | 0 | 1 |
Ae. pembaensis | 0 | 1 | 0 | 0 | 0 | 0 | 1 |
An. gambiae | 0 | 0 | 2 | 0 | 0 | 0 | 2 |
Cx. annulioris | 0 | 0 | 1 | 0 | 0 | 0 | 1 |
Cx. quinquefasciatus | 253 | 1136 | 715 | 211 | 3 | 46 | 2364 |
Cx. rubinotus | 0 | 0 | 0 | 0 | 6 | 0 | 6 |
Cx. univittatus | 1 | 2 | 2 | 0 | 0 | 0 | 5 |
Cx. zombaensis | 16 | 8 | 7 | 12 | 0 | 0 | 43 |
TOTAL | 277 | 1156 | 737 | 529 | 455 | 110 | 3264 |
Blood meal sources
Out of the 161 blood fed female mosquitoes tested by ELISA for host blood meal sources, 91%, (n = 146) were from Culex and the rest Aedes (9%, n = 15) species (Table 5). The samples were tested against four blood meal source/antisera namely: bovine, chicken, goat and human. majority of the samples could not be identified (66.46%) for blood meal sources, but a large portion of the identified had fed on blood of human origin (29.81%) and the least had consumed chicken blood (3.73%). none of the mosquito samples had fed on either goat or bovine. Given that sampling was done in the urban areas where very less or none at all practice farming, livestock couldn’t be found in the homesteads, and this was an expected result. (Table 5). The mosquitoes analyzed comprised of Cx. quinquefasciatus (n = 143), Ae. aegypti (n = 15) and Cx. univittatus (n = 3) (Table 5). Overall, the blood meal preference varied between outdoor and indoor samples with majority at indoor locations (n = 140) and to a lesser extent on outdoors (n = 21). Majority (n = 92) of the indoor sampled mosquitoes tested could not be identified but larger percentage of those identified had fed on human (n = 44), chicken (n = 4). Similarly, for outdoor sampled mosquitoes, a larger portion (n = 13) couldn’t be identified but among those identified, majority had fed on humans (n = 6) and chicken (n = 2)
Majority of the Cx. quinquefasciatus mosquitoes which fed on humans (n = 40) and chicken (n = 4) were captured indoor and to lesser extent in the outdoor environ (humans, n = 3 and chicken, n = 2) (Table 5).
Table 5
Blood meal sources of the blood fed mosquitoes collected in Malindi, Kilifi and Mombasa urban areas
Species | Site | Location | No. tested | Human (%) | Bovine (%) | Goat (%) | Chicken (%) | Unidentified (%) |
Ae. aegypti | Kilifi | outdoor | 4 | 3 (75) | 0 (0) | 0 (0) | 0 (0) | 1 (25) |
indoor | 1 | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 1 (100) |
Malindi | outdoor | 9 | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 9 (100) |
Mombasa | indoor | 1 | 1 (100) | 0 (0) | 0 (0) | 0 (0) | 0 (0) |
Cx. quinquefasciatus | Kilifi | outdoor | 7 | 2 (28.6) | 0 (0) | 0 (0) | 2 (28.6) | 3 (42.9) |
indoor | 19 | 11(57.9) | 0 (0) | 0 (0) | 1 (5.3) | 7 (36.8) |
Malindi | indoor | 73 | 22(30.1) | 0 (0) | 0 (0) | 3 (4.1) | 48 (65.8) |
Mombasa | outdoor | 1 | 1 (100) | 0 (0) | 0 (0) | 0 (0) | 0 (0) |
indoor | 43 | 7 (16.3) | 0 (0) | 0 (0) | 0 (0) | 36 (83.7) |
Cx. univittatus | Malindi | indoor | 1 | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 1 (100) |
Mombasa | indoor | 2 | 1 (50) | 0 (0) | 0 (0) | 0 (0) | 1 (50) |
Total | | | 161 | 48(29.8) | 0 (0) | 0 (0) | 6 (3.7) | 107 (66.5) |
For Aedes aegypti, 15 mosquito samples were tested for host blood meals. Eighty-seven percent 87% (n = 13) were captured outside houses while 13% (n = 2) were trapped inside houses. Out of the 15 Aedes aegypti, 73% (n = 11) had fed on unidentified hosts while the rest (27%, n = 4) had feed on humans (3 outdoors & 1 indoors).
Three Cx. univittatus were tested for host blood meal. All of these had sourced their blood meal inside houses. Majority (67%, n = 2) of these had fed on unidentified hosts while the rest (33%, n = 1) fed on humans.
Arboviruses diversity in mosquitoes
Out of 259 pools screened against the three viral genera and 11.58% pools tested positive. Overall, the pools consisted of 129 Ae. aegypti pools and 130 Cx. quinquefasciatus pools. The overall positive pools (n = 30) were only positive to Flavivirus and none for either Phlebovirus or Alphavirus. The Ae. aegypti had a significantly higher (ᵡ2, (df = 1, n = 30) = 18.4398, P = 0.000) proportion of virus positive pools (87%, n = 26) compared to Cx. quinquefasciatus (13%, n = 4).
Ae. aegypti mosquitoes had 129 (60 females and 69 males) pools screened, 20.16% (n = 26) of the pools turned to be positive for flavivirus. There was site to site variation in terms of flavivirus positivity in the mosquito pools in three sites (ᵡ2, (df = 2 n = 30) = 14.2292, P = 0.001). In Kilifi, 18 pools (5 for females & 13 for males) of Aedes aegypti mosquitoes tested positive for flavivirus. In Mombasa, only three pools were positive comprising of Aedes aegypti mosquitoes only (1 pool for female and 2 for males). In Malindi, five pools of Aedes aegypti mosquitoes tested positive (all female pools). There was no significant difference between male and female positivity (ᵡ2, (df = 1 n = 30) = 0.2697, P = 0.604) (Table 6).
Culex quinquefasciatus mosquitoes had only 4 pools which tested positive, 1 pool in Kilifi, 3 in Malindi and none from Mombasa (Table 6). All flavivirus positive samples were negative for dengue virus.
Table 6
Total number of pools positive for flavivirus in the Kilifi, Mombasa and Malindi
| | | Total number of pools (positive) |
Species | Site | Sex | BG | LT | Larvae | soil sample |
Ae. aegypti | Kilifi | F | 8 (3) | 1 (1) | 5 (1) | 5 (0) |
M | 14 (5) | 4 (3) | 6 (2) | 8 (3) |
Malindi | F | 13 (3) | 3 (0) | 6 (0) | 6 (2) |
M | 11 (0) | 2 (0) | 7 (0) | 4 (0) |
Mombasa | F | 4 (1) | 3 (0) | 5 (0) | 0 (0) |
M | 3 (0) | 3 (1) | 8 (1) | 0 (0) |
| Sub total | | 53 (12) | 16 (5) | 37 (4) | 23 (5) |
Cx. quinquefasciatus | Kilifi | F | 6 (1) | 9 (0) | 4 (0) | 0 (0) |
M | 8 (0) | 6 (0) | 2 (0) | 0 (0) |
Malindi | F | 1 (0) | 35 (3) | 0 (0) | 0 (0) |
M | 2 (0) | 17 (0) | 0 (0) | 0 (0) |
Mombasa | F | 2 (0) | 23 (0) | 1 (0) | 0 (0) |
| M | 3 (0) | 10 (0) | 1 (0) | 0 (0) |
| Sub total | | 22 (1) | 100 (3) | 8 (0) | 0 (0) |
| Grand total | | 75 (13) | 116 (8) | 45 (4) | 23 (5) |