Anopheles gambiae complex contains the most efficient malaria vectors globally. Such vectors include Anopheles gambiae s.s and An. arabiensis [1, 2]. Anopheles arabiensis has been associated with malaria transmission in semi arid, arid and high altitude areas [3, 4]. This species may be zoophilic and/or exophilic in nature and sometimes preferring to feed on animals (especially bovines) relative to humans [5, 6]. Indeed, in most of the behavioural studies, An. arabiensis have shown higher preference for domestic bovines as compared to other domestic animals [5–10].

Different mosquito sampling techniques are in use such as human landing catch which is highly facing ethical barriers, complex odour baited traps which needs trained staffs and CDC light traps which needs recharged power system to operate for long ; however most of them have a limited use thus new techniques are needed to complement or replace the existing ones. There is a high interest in developing other tools due to ethical concerns against human landing catch (HLC) which has been used as a gold standard [11–13]. Odour baited traps are receiving keen attention as they do not involve human as a subject and most of the time they require minimum supervision and operations during the night. Attractive odours have been shown to increase trap efficiency as female mosquitoes mainly use odour in locating their hosts [11, 14]. Urine from different animals has successfully been used as attractants for different insects [15]. Cattle urine has for long time been proven as a good source of attraction for tsetse flies [15, 16]. Indeed, incorporation of host odour, artificial or synthetic (cattle urine or sebum, chemical compounds) in the traps has improved the control of tsetse flies. Horse's urine has also proved a strong attraction towards tabanids [17]. Recently a study done in lower Moshi, Tanzania suggested an existing potential for using fresh cattle urine as an attractant in sampling adult An. arabiensis mosquitoes [11]. However, that study did not determine the duration at which urine could remain efficient as an attractant after application to the resting boxes, and effect of the aging of urine on its attraction. This study was therefore aimed at responding to such questions. Three stages of decaying cattle urine were evaluated at lower Moshi rice irrigation schemes, an area ought to have 95% of the anopheline population formed by An. arabiensis [18].

This study tested aging urine to indicate the possibility of its use for sampling An. arabiensis. Kweka et al. [11] proved the usefulness of fresh cattle urine for sampling An. arabiensis under field settings. Fresh urine has also been used for field collection of tsetse flies [20–22]. Information on the length of time at which cattle urine remains attractive after application remains necessary and will assist in determining re-treatment rates and improvement.

Although not as much as with fresh urine, the present study has demonstrated collection of high proportion of mosquitoes using traps re-treated 3 and 7 days after initial application. Out of 7871 mosquitoes which were collected, 21.6% and 20% were from traps re-treated with urine every 3 and 7 days respectively. 8.6% was collected from traps not treated with urine. This signifies that cattle urine could still be useful for collecting mosquitoes several days post its initial application. In studies where urine was evaluated against tsetse no flies were collected in the traps with aging urine [17, 21, 22].

The proportion of mosquitoes collected indoor was ~2 folds greater than that collected outdoor. Anopheles arabiensis is 25 to 50 percent exophilic in nature depending on the locality [6], thus we would expected to have more of them collected outside than inside houses. This might be partially due to absence of wind blows indoors hence causing the boxes to be wet for long time with concentrated odour than those boxes outdoor which may be affected by wind blow. Despite the difference in indoor and outdoor catches, it is still worthy saying that resting boxes can be used for indoor and outdoor sampling of An. arabiensis. This suggests the potential of incorporating synthetic or biological insecticides in cattle urine baited traps (lure and kill) to target outdoor or indoor resting mosquitoes. However, for the outdoor targeting, box traps will require a water proof casing to ensure their use during rains. In addition to lure and kill, using urine odours may increase the efficiency of the existing tools without using humans as baits. This will ensure determination of malaria vectors and parasites dynamics using simple and ethically acceptable tools rather than human landing catch. Study done in northern Tanzania showed that CDC light trap were by 22.64% more efficient than clay pots [23]. Coupling pots with cattle urine odour will probably increase their efficiency thus offering additional traditional sampling tools as effective as CDC light trap.

The resting box traps captured all range of mosquitoes with different abdominal status, unfed (12.3%), freshly fed (4.1%), semi-gravid (34.2%) and gravid (49.4%). However, a large proportion was made of semi-gravid and gravid mosquitoes. Similarly, in the study where pots were used, many sampled mosquitoes were semi-gravid and gravid [23]. After feeding, mosquitoes are badly looking for conducive places to rest, digest blood and lay eggs. Conditions within the resting boxes were probably good enough thus many fed mosquitoes remained. It could be that, almost equal numbers of unfed mosquitoes were attracted by the odour but escaped to continue scavenging for blood source hence lower density of unfed An. gambiae s.l.

Cheap but effective sampling tools are obviously required for sustainable use in the malaria endemic regions where coincidentally >85% of the people are poor and live in rural areas. The present study used resting boxes and black cotton cloth material which can be obtained locally and at relatively low cost. The boxes are portable and of small sized and can be placed in any place without disturbing house owners. For routine surveillance not many areas would always afford to use tools like Magnetic traps with synthetic odours although such tools have been used successfully [24–26]. Other sampling tool such as odour baited entry trap have shown to be effective and protective to human participants but needs to have several machines, tents and a security [7, 25].

Fresh and decaying/aging cattle urine can be used to odour bait resting boxes thus increasing their efficiency in sampling An. arabiensis, more so for semi-gravid and gravid mosquitoes. However, evaluation in low density seasons of An. arabiensis in different ecological settings will yield more information to aid in the rational use of resting boxes as sampling tools. Later odour baited resting boxes can be standardized to replace human landing catch as a sampling tool for An. arabiensis and other malaria vector species.

The authors declare that they have no competing interests.