Acta Tropica 85 (2003) 71 /82 www.parasitology-online.

com

The phenology of malaria mosquitoes in irrigated rice elds in Mali

E. Klinkenberg a,b, W. Takken b,1, F. Huibers a, Y.T. Toure c
a

Irrigation and Water Engineering Group, Wageningen University and Research Centre, Nieuwe Kanaal 11, 6709 PA, Wageningen, The Netherlands b Laboratory of Entomology, Wageningen University and Research Centre, PO Box 8031, 6700 EH, Wageningen, The Netherlands c Departement depidemiologie des affections parasitaire, Universite du Mali, PO Box 1805, Bamako, Mali Received 16 August 2001; received in revised form 8 October 2002; accepted 22 October 2002

Abstract A field study was carried out in the large-scale rice irrigation scheme of the Office du Niger in Mali to investigate the relation between anopheline mosquito larval development and small-scale differences in irrigation practices, such as water level, irrigation application and irrigation frequency. The objective of the study was to find out if water management can be used as a tool for vector control to reduce the malaria transmission risk. Larvae of Anopheles gambiae s.s ., the main malaria vector in the study area, developed mostly in the first 6 weeks after transplanting the rice. During rice development, a succession of anopheline species was observed. This was associated with a marked decrease in light intensity reaching the water surface as plant height increased. Minor differences in water management resulted in noticeable variations in larval densities and species composition. A. gambiae s.s. larvae were most abundant during the early growing stages and almost absent in a closed rice crop. Due to improper drainage after harvest, A. gambiae s.s. breeding was soon re-established in fields where small pools of water were retained. The results suggest that larval mosquito habitats in the Office du Niger can be significantly reduced by water management, simultaneous planting and harvesting and proper drainage of fallow fields. # 2002 Elsevier Science B.V. All rights reserved.
Keywords: Anopheles larvae ; Malaria vectors; Irrigation; Mosquito population dynamics; Mali

1. Introduction Irrigated agriculture and especially rice cultivation has a strong association with malaria vectors as many anophelines lay their eggs in the flooded
1 Corresponding author. Tel.: '/31-317-484-652; fax: '/31317-484-821 E-mail address: willem.takken@wur.nl (W. Takken).

rice fields and the irrigation and drainage channels (Lacey and Lacey, 1990; Ijumba and Lindsay, 2001). The introduction of irrigation has substantially prolonged the season during which adult malaria vectors can be found in areas with seasonal rainfall as mosquitoes have access to larval habitats perennially. However, in Africa, contradictory results are reported on the effect of irrigation on malaria risk. In general, it is stated

0001-706X/02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 1 - 7 0 6 X ( 0 2 ) 0 0 2 5 4 - 1

72

E. Klinkenberg et al. / Acta Tropica 85 (2003) 71 /82

that development of irrigation increases the prevalence of malaria. In the Benoue Valley of northern Cameroon, malaria incidence increased after large-scale irrigation development following completion of the Lagdo dam (Robert et al., 1992). In the Tigray region of Ethiopia, the overall incidence of malaria was seven times higher in villages close to dams compared with controls (Ghebreyesus et al., 1999). In contrast, the introduction of irrigation systems decreased the prevalence of malaria in the Vallee du Kou in Burkina Faso (Robert et al., 1988; Boudin et al., 1992). The reasons for these contradictory results following introduction of rice irrigation, may be numerous, such as improved income levels, better housing and medical care in case of malaria decrease (Ijumba and Lindsay, 2001). Alternatively, malaria increases are usually ascribed to increased vector densities and changes in vector phenology as a result of the irrigation. However, increased vector densities have sometimes coincided with reduced sporozoite rates and, hence, reduction in malaria prevalence (Dietz 1988; Robert et al., 1991, 1992). Most research on irrigation and malaria is focused on the prevalence of the disease within the human population close to the irrigation scheme. Few studies address the direct effect of the irrigation system, in particular water management, on the vectors themselves, species of the mosquito genus Anopheles . Much research on this subject comes from Asia and dates from the first half of the 20th century when water management was successfully used for malaria control (Takken et al., 1990). In Asia, the anopheline species responsible for malaria transmission are different from those in Africa and they have different ecological niches. It, therefore, seems likely that different water management strategies are necessary for malaria vector control in the two continents. In China (Pao Lingh Luh, 1984) and Indonesia (Snellen, 1990) intermittent irrigation is successfully practised for mosquito control. In Africa, this method of mosquito control is not necessarily successful and may even increase breeding of Anopheles gambiae Giles, the main malaria vector on the continent. This species prefers shallow sunlit pools for breeding and these

are often created after drainage of the fields. Grainger (1947) reported no effect of intermittent irrigation on mosquito breeding in rice fields in western Kenya. In central Kenya, a seasonal population increase of A. gambiae s.l. , mostly A. arabiensis Patton, and A. pharoensis Theobald was associated with periodic flooding of irrigated rice fields (Mukiama and Mwangi, 1990; Mwangi and Mukiama, 1992). In the same area, Mutero et al. (2000) found that intermittent irrigation favours egg laying by A. arabiensis Patton but reduced the larval survival rate. Here we report on a study undertaken to investigate the effect of water management and rice cultivation on malaria vectors in the Office du Niger in Mali to determine to what extent field water management can be considered as a tool for vector control.

2. Study area and study sites 2.1. Study area The Office du Niger is located in the inner delta of the Niger River, approximately 350 km north of Bamako (Fig. 1). It is a large-scale gravity irrigation scheme that was built in the 1930s. The irrigation system consists mostly of unlined earthen channels fed by water from the river Niger. The irrigation potential of the area is about 960 000 ha, but only 60 000 ha has been developed and expansion of the irrigated area no longer takes place (DGIS, 1992). The main intake of the irrigation system is at the Markala dam in the river Niger, from where water is diverted to five irrigated rice zones and one sugarcane plantation, located 60 /120 km from the water intake (Fig. 1). Rice is cultivated according to a single or double cropping calendar (Toure et al., 1997). Where only a single crop is grown, this coincides with the rainy season, May /September. Where there is double cropping, the first crop is planted in March, harvested in June and then followed by a second crop. The climate in the area is Sudano /Sahelian with one rainy season. During the year, three seasons can be distinguished: a warm dry season (March / May), a rainy season (June /September) and a cool

E. Klinkenberg et al. / Acta Tropica 85 (2003) 71 /82

73

ences in water supply. In general, fields at the end of irrigation channels experience more water shortage than fields at the beginning of a channel. Four fields with a double cropping calendar and six fields with a single cropping calendar were selected. The fields with double cropping were in the second half of rice development at the start of the study period (beginning of May) as the rice had been transplanted in March. These are referred to as early rice. The fields were harvested in June /July and it was expected that they would be replanted soon after harvesting. However, these fields lay fallow for several weeks after harvesting and were not replanted before the end of the study period. In the fields with a single crop, rice was transplanted in May /June to be harvested in September. These fields are referred to as late rice.

3. Material and methods 3.1. Mosquito collection

Fig. 1. Map of the Ofce du Niger (after Toure et al., 1997) and location of the study site.

dry season (October /February). Average rainfall is 400 /600 mm per year. Since rainfall is unreliable, rice cultivation is often not possible without irrigation. The total population of the Office du Niger was estimated at 150 000 persons in 1991 (DGIS, 1992). In this region of Mali malaria is among the main causes of morbidity and mortality (WARDA, 1997). Next to malaria, schistosomiasis and water-borne diseases are other diseases associated with irrigation that are highly prevalent in the area. 2.2. Study sites Ten fields for larval studies were selected near the village of Niono Coloni, which is adjacent to the village of Niono and is located in one of the five rice growing zones (Fig. 1). Fields were selected at the beginning and end of the irrigation channels to include fields with the largest differ-

Between May and July 1999, mosquito larvae were collected with a 240 ml dipper after Service (1993). Across each field, two different transects were made daily guided by a length of rope. A transect always included part of the edge of the field. Five measuring points were selected along each transect and within one square meter of each measuring point 10 dips were taken, i.e. 100 dips per field. The location of the transect was changed each day to prevent influence on larval numbers by dipping each day at the same point. Mosquito collections were carried out in five fields per day, i.e. the same field was sampled every other day. Larval collections were counted and identified. Anophelines were identified to species, culicines to genus only. Larvae were identified by the naked eye and where necessary with a magnifying lens or microscope or reared till a later stage for easier identification. Pupae were reared till the adult stage for species identification. Weekly 10 /15 A. gambiae larvae from each field were collected and stored in Carnoys fixative (three parts pure ethanol to one part pure glacial acetic acid) for PCR analysis later after Scott et al. (1993).

74

E. Klinkenberg et al. / Acta Tropica 85 (2003) 71 /82

3.2. Water measurements Water levels were measured daily in all ten fields by reference to a fixed measuring point (iron bar fixed in cement) that was installed in each field. Fluctuation in water levels was determined by measuring the distance between the water surface and the top of the bar. For each field an estimate was made of how much of the field was covered by water, using seven categories, from completely dry to 80 /100% water coverage. For each measuring point it was noted how much of the square meter was covered by water and, if pools were formed, the number and average size of the pools was noted. 3.3. Plant measurements The height and density of the rice plants and the amount of light reaching the water surface were measured weekly. The height of the rice plants was measured at three points in the field and the mean height was considered representative for the entire field. Two measures of density were used, the number of plant clusters per m2 and the number of tillers per cluster. The average number of plant clusters per m2 was obtained by counting at three points in the field. The average number of tillers per plant cluster was estimated by counting the number of tillers in ten clusters. The light intensities above the vegetation and just above the water surface were measured with a photometer. The ratio of these two values was used as an indicator for the amount of light reaching the water surface. 3.4. Statistical analysis Data were statistically analysed using SPSS for (version 8.0). Regression analysis and (M)ANOVA tests were used to determine if differences were statistically significant. Larval data were log-transformed to improve normality of distribution and homogeneity of variances. As data came from different fields a dummy variable was inserted into the regression model to account for differences between fields. Results are reported as significant if P B/0.05 unless stated otherwise.
WINDOWS

4. Results The total rainfall for the period 1st May /27th July 1999 was 251.5 mm. In this period, average daily maximum temperatures ranged from 30.6 to 43.0 8C and average daily minimum temperatures from 23.0 to 29.0 8C. The maximum relative humidity ranged from 33 to 76% and the minimum relative humidity from 15 to 52% (data Meteorological Station, Niono). 4.1. Anopheline species composition in the study area In the rice fields under study four anopheline species were found: A. gambiae s.l. , A. pharoensis Theobald, A. funestus Giles and A. rufipes (Gough). About 65% of all larvae collected were A. gambiae s. l. PCR identification of 186 larvae of A. gambiae s.l. gave a positive result for 125 specimens. Of these, 98.4% were A. gambiae s.s. , and 1.6% were A. arabiensis . The former were not further classified into the different chromosomal forms (Favia et al., 1997; Della Torre et al., 2001). There was no specific spatial distribution of the anopheline species within the rice fields studied. There was, however, a large variation in species composition between fields, both in total numbers (Fig. 2) as well as over time. 4.2. Anopheline development over time From field observations it was estimated that larval development in the Niono area takes about 8/9 days from egg to adult. This was deduced from a field that was inundated on 31st May and in which stage IV larvae and pupae appeared on 7th June (data on composition of larval stages over time not shown here). Regarding larval development the fields could be classified into two main groups: fields with a late rice crop Fig. 3(a) and fields with an early rice crop Fig. 3(b) (see above). In the fields with a late rice crop the number of larvae gradually increased from the time of transplanting and decreased after 35/40 days (5 /6 weeks), although the timing of this point of decrease varied between fields Fig. 3(a). The fields with an early rice crop, which were

E. Klinkenberg et al. / Acta Tropica 85 (2003) 71 /82

75

Fig. 2. Anopheline species composition for each eld as the percentage of total specimens collected on that eld. Fields 1 /4, early rice, second development period; Fields 5 /10, late rice, rst development period.

studied from about 80 days after transplanting, had more constant and lower larval densities Fig. 3(b), with the exception of field 1 that experienced a sudden increase and more fluctuation in larval densities. One field was poorly levelled and one corner of it often became dry during the time of the study. The larval densities in fields in the second half of rice development (early rice crop) were significantly lower (P B/0.001) than in fields in the first half of rice development (late rice crop).

4.3. Anopheline development under inuence of developing rice During development of the rice, a succession of anopheline species occurred. A. gambiae and A. pharoensis dominated the first 6 /8 weeks of the rice cycle, whereafter, their numbers sharply declined and dominance was gradually taken over by A. rufipes and A. funestus (Fig. 4). However, the densities of A. gambiae during the early period of rice growing were never equalled by these of A. rufipes or A. funestus . In the fields with

an early rice crop (fields 1/4), that were in the second half of rice development during the study period, the population was dominated by A. rufipes while A. funestus remained relatively rare. Stepwise regression analysis indicated that the percentage of light reaching the water surface explained almost 50% of the variation found (Table 1). The average weekly larval densities of A. gambiae and A. pharoensis combined, and A. rufipes and A. funestus combined, are shown in Fig. 4 with the percentage of light reaching the water surface. These data show that when this percentage was less than 80%, the switch to dominance of the latter group occurred. Anopheline development during the second half of rice development, from day 60 after transplanting onwards, was studied in fields that were planted early in the growing season, well before the onset of the rains. Rice plants were well developed at this time and the percentage of light reaching the water surface was between 60 and 80%. The densities of the different Anopheles species in these fields from day 100 after transplanting onwards are shown in

76

E. Klinkenberg et al. / Acta Tropica 85 (2003) 71 /82

Fig. 3. (a) Larval densities in the total anopheline population (#larvae/100 dips) over time (days after transplanting) in the elds 5 /10 with a late rice crop. (b) Larval densities in the total anopheline population (#larvae/100 dips) over time (days after transplanting) in the elds 1 /4 with an early rice crop.

Fig. 5. The vertical line indicates the time of harvesting, when the light reaching the water surface suddenly increased to 95/100%. At this point A. gambiae quickly re-occupied these fields, provided that water remained standing on the now fallow fields. Most fields were not drained after harvest and were not immediately prepared for the next cultivation cycle. Such fields lay fallow for up to several weeks containing hundreds of small pools, providing excellent breeding sites for A. gambiae.

4.4. Anopheline densities and water management The water measurements were used to assign fields to different categories of water management (Table 2 and Klinkenberg et al., 2002). There were small apparent differences in larval densities between the different categories, but they were not significant (Table 2). Fields in category B had a lower average larval density than those in category A. Fields 5 and 8 can be regarded as a separate group (category C), which had the highest

E. Klinkenberg et al. / Acta Tropica 85 (2003) 71 /82

77

Fig. 3 (Continued)

average larval density but this was mainly due to one peak occurrence and the range of values was comparable to that of category B fields.

5. Discussion The rice fields in the Office du Niger area appear to be excellent breeding habitats for several anopheline species, with A. gambiae s.s. being dominant. As the latter species outnumbers the other vector species, it seems that, at least in the rice fields during the growing season, most of the malaria is transmitted by A. gambiae s.s. , A. funestus , which is as efficient a vector as A. gambiae s.s. , was found only in low densities and only during the second half of the rice cultivation when the plants provided shade. Although anopheline larvae were frequently encountered in the small irrigation channels leading to the rice fields and other potential breeding habitats, such as seepage areas and pools, we did not quantitatively investigate anopheline production in these breeding sites. The relative surface area of the rice fields

was very much larger than that of the channels, however, and we can consider the irrigated fields to be the principle breeding sites. Development of A. gambiae occurred primarily in the first 6 weeks after transplanting and rice was not transplanted simultaneously in all fields, so that at any time during the study period there were large numbers of fields providing ideal habitats for A. gambiae . This species was the most abundant of the anophelines found and is an efficient vector throughout Africa. Therefore, we assume that, at least during the rice cultivation season, most of the malaria in the study area was transmitted by this species. Of the other anophelines, that were found in much lower densities, only A. funestus can be considered an important vector, based on data from elsewhere in Africa (Gillies and Coetzee, 1987). The vector status of A. pharoensis and A. rufipes in north /central Mali is uncertain, but it is assumed to be secondary. A. pharoensis bites humans as well as animals but is rarely found in human bait catches in the Niono area (Y.T. Toure, unpublished data). A. pharoensis is involved in malaria transmission in the Senegal River delta

78

E. Klinkenberg et al. / Acta Tropica 85 (2003) 71 /82

Fig. 4. Average weekly larval densities of pooled A. gambiae s.l. and A. pharoensis and pooled A. rupes and A. funestus plotted together with the percentage of light reaching the water surface.

Table 1 Results of regression analysis to investigate the percentage of variation found explained by different agronomic parameters for elds with a late rice crop Model number Data of fields Dependent variable (log) Variables entered in the model Variation explained (%) 49.7 57.9 61.8 44.8

1 2 3 4

6 /10 6 /10 6 /10 6 /10

Anophelines A gambiae s.l. A. gambiae plus A. pharoensis A. funestus and A. rufipes

Percentage light, tillers Percentage light, height, stage, tillers Percentage light, height, stage, tillers Height, tillers

Only the best fits are shown. %light, percentage of light reaching the water surface; tillers, average number of tillers per pocket; height, average height of the rice vegetation; stage, developmental stage of the rice.

E. Klinkenberg et al. / Acta Tropica 85 (2003) 71 /82

Fig. 5. Anopheline species composition in relation to water coverage on four different elds from day 100 after transplanting onwards.

79

80 Number of days after which larval densities decrease

E. Klinkenberg et al. / Acta Tropica 85 (2003) 71 /82 Category A, fields with a water level between 10 and 20 cm and frequent irrigation (6 /8 day interval) with relatively small applications (20 /60 mm); category B, fields with a higher water level (20 /30 cm), which receive a large irrigation application at the start of the season and irregularly small applications afterwards; category C, fields that were used as a seedbed before transplanting. a Field had one large application to start with as pre-irrigation or after field has fallen dry. b These fields have one large application at the start of the season and irregularly small applications afterwards. c There occurred one large peak density. (Table is after Klinkenberg et al., 2002).

(Carrara et al., 1990) and in Cameroon (Robert et al., 1992). A. rufipes prefers animals to man as a host. Recent studies on malaria transmission in the Office du Niger indicate that A. gambiae is indeed the most important vector (Dolo et al., 1999) for this area. As larval populations of A. gambiae sharply declined or even disappeared during the second stage of the rice crop, we suggest that manipulating the planting time of the rice could be a potential strategy for suppression of malaria vectors in this area. It was interesting to find that /98% of the A. gambiae found were A. gambiae s.s. This was also found by a previous study in the same area (Dolo et al., 1999). A. gambiae s.s. is considered the most efficient malaria vector due to its intensely anthropophilic and endophilic character coupled with a relatively high longevity (Toure et al., 1994). In the Gambia and Senegal, situated to the West and at the same latitude as the study area, A. arabiensis may be locally the dominant malaria vector especially in the dry season (Fontenille et al., 1997; Lindsay et al., 1991; Lemasson et al., 1997). It is possible that rice cultivation has favoured A. gambiae s.s and that previously A. arabiensis occurred in higher ratios compared with that found in this study. Elsewhere in Mali (Toure et al., 1994) and in neighbouring Burkina Faso, in particular during the dry season, A. arabiensis is the dominant species (Costantini et al., 1996). Toure et al. (1996) also found A. gambiae s.s. to be dominant in both the dry and the rainy season in the village of Douna in the North Sudan Savanna zone of Mali. Colluzi, 1992 emphasised the effects of human activities, particularly irrigation, on development of anopheline vectors and species composition in West Africa. Indeed, it has been suggested that agriculture was one of the reasons for rapid speciation of the A. gambiae complex in Mali (Favia et al., 1997). The dominance of A. gambiae and A. pharoensis was replaced by a dominance of A. rufipes and A. funestus in the course of the growing season. Other studies in Burkina Faso (Robert et al., 1988) and Kenya (Chandler and Highton, 1975, 1976) found a similar breeding pattern. The succession of species has important consequences for malaria transmission as A. gambiae, the main malaria

Occurrence of large peak densities Entomological characteristics Table 2 Characteristics of irrigation water management and larval development in elds with a late rice crop (#larvae/100 dips) average (25 /75% percentile) Irrigation interval (days) Category Water level (cm) Irrigation application (mm) Field Water management characteristics

6 9 10 5 8 7

A A A C (A) C (B) B

10 /20 20 /60a 20 /60 10 /20 10 /20 20 /60 10 /25 20 /60a Variable (10 /25) Variableb (70 /130) 15 /30 Variableb (30 /130)

6 /8 6 /8 6 /8 variable variable (3 /14) 9 /12

99 (45 /121) 75 (39 /113) 89 (52 /130) 109 (30 /134)c 105 (25 /136)c 67 (14 /83)

No No No Yes Yes No

35 35 37 22 25 48

E. Klinkenberg et al. / Acta Tropica 85 (2003) 71 /82

81

vector, almost disappeared from the rice fields at this point. A. funestus , a vector of similar efficiency as A. gambiae s.s. , only occurred in the second half of rice development but remained rare. From this study, it is concluded that the timing of starting of cultivation was probably the most important factor in maintaining the continuous presence of the malaria vector A. gambiae in the study area (Klinkenberg et al., 2002). This was mainly caused by the non-synchronous method of cultivation. Farmers work in rotation because they can not all draw water from the system at the same time. In this way a habitat ideal for A. gambiae breeding, shifts from field to field and the mosquitoes can shift correspondingly. A possible solution to break this pattern of continuous mosquito breeding would be a synchronisation of transplanting in large blocks at least 5/7 km apart, i.e. further apart than the maximum flying distance of anophelines. Furthermore, the practice of leaving the fallow fields inundated allows the creation of new anopheline breeding sites, which would not be present if the fallow fields were properly drained. The continuous breeding of anophelines has important implications for the malaria risk in the study area. This leads to an additional malaria risk in the dry season, which is not present in nonirrigated areas. In the Sahelian region of Senegal the prevalence of malaria is normally restricted to the rainy season (Fontenille et al., 1997), and this is likely to be the case in Mali as well in nonirrigated areas.

This study was supported by the Wageningen Agricultural University Fund. The authors wish to acknowledge the two anonymous reviewers for their valuable suggestions for improvement of the manuscript.

References
Boudin, C., Robert, V., Carnevale, P., Ambroise-Thomas, P., 1992. Epidemiology of Plasmodium falciparum in a rice eld and a Savanna area in Burkina Faso. Comparative study on the acquired immunoprotection in native populations. Acta Trop. 51, 103 /111. Carrara, G.C., Petrarca, V., Niang, M., Coluzzi, M., 1990. Anopheles pharoensis transmission of Plasmodium falciparum in the Senegal River delta. West Afr. Med. Vet. Entomol. 4, 421 /424. Chandler, J.A., Highton, R.B., 1975. The succession of mosquito species in rice elds in the Kisumu area in Kenya, and their possible control. Bull. Ent. Res. 65, 295 /302. Chandler, J.A., Highton, R.B., 1976. The breeding of Anopheles gambiae Giles (Diptera; Culicidae) in rice elds in the Kisumu area of Kenya. J. Med. Entomol. 13, 211 /215. Colluzi, M., 1992. Malaria vector analysis and control. Parasitol. Today 8, 113 /118. Costantini, C., Li, S.G., Della Torre, A., Sagnon, N., Coluzzi, M., Taylor, C.E., 1996. Density, survival and dispersal of Anopheles gambiae complex mosquitoes in a West African Sudan Savanna village. Med. Vet. Entomol. 10, 203 /219. Della Torre, A., Fanello, C., Akogbeto, M., Dossouyovo, J., Favia, G., Petrarca, V., Coluzzi, M., 2001. Molecular evidence of incipient speciation within Anopheles gambiae s.s. in West Africa. Insect Mol. Biol. 10, 9 /18. DGIS, 1992. La riziculture paysanne a lOfce du Niger, Mali, ` 1979 /1991. Cooperation au developpement des Pays-Bas. Ministere des affaires etrangeres, La Haye, Pays-Bas. ` ` Dietz, K., 1988. Density-dependence in parasite transmission dynamics. Parasitol. Today 4, 91 /97. Dolo, G., Doa, A., Sogoba, N., Boaure, M., Sangar, D., Bagayoko, M., Traore, S.F., Niare. O., Sagara, I., Dicko, A., Sissoko, M.S., Sissoko, M., Doumbo, O.K. Toure,Y.T., 1999. Impact of irrigated rice cultivation on malaria transmission in Niono, Mali. Abstract and presentation MIM congress, Durban, South Africa, March 1999. Favia, G., Della Torre, A., Bagayoko, M., Lanfrancotti, A., Sagnon, N.F., Toure, Y.T., Coluzzi, M., 1997. Molecular identication of sympatric chromosomal forms of Anopheles gambiae and further evidence of their reproductive isolation. Insect Mol. Biol. 6, 377 /383. Fontenille, D., Lochouarn, L., Diagne, N., Sokhna, C., Lemasson, J.-J., Diatta, M., Konate, L., Faye, F., Rogier, C., Trape, J.-F., 1997. High annual and seasonal variations in malaria transmission by anophelines and vector species

Acknowledgements We wish to express our gratitude to all the farmers of the village of Niono Coloni who allowed us to work in their elds. Furthermore, we are grateful to Dr Magaran Bagayoko and all other members of the Malaria Research and Training Centre (MRTC) in Bamako for their assistance and hospitality. Also the Institut Economique Rurale (IER) in Niono, under the leadership of Dore Guindo, is gratefully acknowledged for providing workspace, technical assistance and advice. We especially want to thank Sekou Sala Guindo of the EIR for his assistance in the eld.

82

E. Klinkenberg et al. / Acta Tropica 85 (2003) 71 /82 in China. In: Environmental management for vector control in rice elds, FAO Irrigation and drainage paper 41, FAO Rome. Robert, V., Ouari, B., Ouedraogo, V., Carnevale, P., 1988. Etude ecologique des Culicidae adultes et larvaires dans une riziere en Valee du Kou, Burkina Faso. Acta Trop. 45, 351 / ` 359. Robert, V., Petrarca, V., Coluzzi, M., Boudin, C., Carnevale, P., 1991. Etude des taux de parturite et dinfection du complexe Anopheles gambiae dans la riziere de la vallee du ` Kou, Burkina Faso. In: Robert, V., Chippeaux, J.P., Diomande, L., et al. (Eds.), Le paludisme en Afrique de lOuest. Etudes entomologiques et epidemiologiques en zone rizicole et en milieu urbain. Editions de lORSTROM, Paris. Robert, V., Broek, A.van den, Stevens, P., Slootweg, R., Petrarca, V., Coluzzi, M., Goff, G.L.e, Di Deco, M.A., Carnevale, P., 1992. Mosquitoes and malaria transmission in irrigated rice-elds in the Benoue valley in northern Cameroon. Acta Trop. 52, 201 /204. Scott, J.A., Brogdon, W.G., Collins, F.H., 1993. Identication of single specimens of the A. gambiae complex by the polymerase chain reaction. Am. J. Trop. Med. Hyg. 49, 520 /529. Service, M.W., 1993. Mosquito Ecology, Field Sampling Methods, second ed.. Elsevier Applied Science, London/ New York. Snellen, W.B., 1990. Success and failure of malaria control through species sanitation */some practical examples. Wageningen Agric. Univ. Papers 7 (90), 81 /128. Takken, W., Snellen, W.B., Verhave, J.P., Knols, B.G.J., Atmosoedjono, S., 1990. Environmental measures for malaria control in Indonesia-an historical review on species sanitation. Wageningen Agric. Univ. Papers 7 (90), 167. Toure, Y.T., Petrarca, V., Traore, S.F., Coulibally, A., Maiga, H.M., Sankare, O., Sow, M., Di Deco, M.A., Coluzzi, M., 1994. Ecological genetic studies in the chromosomal form Mopti of Anopheles gambiae s.str. in Mali, West Africa. Genetica 94, 213 /223. Toure, Y.T., Traore, S.F., Sankare, O., Sow, M.Y., Coulibaly, A., Esposito, F., Petrarca, V., 1996. Perennial transmission of malaria by the Anopheles gambiae complex in a north Sudan Savanna area of Mali. Med. Vet. Entomol. 10, 197 / 199. Toure, A., Zanen, S., Kone, N., 1997. La restructuration de lOfce du Niger, contribution ARPON III, Cooperation neerlandaise, Ofce du Niger 1997. WARDA (1997). Second technical and nancial progress report (May 1996 /April 1997).

competition in Dielmo, a holoendemic area in Senegal. Am. J. Trop. Med. Hyg. 56, 247 /253. Ghebreyesus, T.A., Haile, M., Witten, K.H., Getachew, A., Yohannes, A.M., Yohannes, M., Teklehaimanot, H.D., Lindsay, S.W., Byass, P., 1999. Incidence of malaria among children living near dams in northern Ethiopia: community based incidence survey. Br. Med. J. 319, 663 /666. Gillies, M.T., Coetzee, M., 1987. A supplement to the Anophelinae of Africa south of the Sahara (Afrotropical region). South African Institute of Medical Research, Johannesburg. Grainger, W.E., 1947. The experimental control of mosquito breeding in rice elds in Nyanza Province, Kenya, by intermittent irrigation and other methods. East Afr. Med. J. 24, 16 /22. Ijumba, J.N., Lindsay, S.W., 2001. Impact of irrigation on malaria in Africa: paddies paradox. Med. Vet. Entomol. 15, 1 /11. Klinkenberg, E., Huibers, F., Takken, W., Toure, Y.T., 2002. Water management as a tool for malaria mosquito control, the case of the Ofce du Niger in Mali. Irrigation and Drainage Systems 16, 201 /212. Lacey, L.A., Lacey, C.M., 1990. The medical importance of riceland mosquitoes and their control using alternatives to chemical insecticides. J. Am. Mosq. Control Assoc. (Suppl.) 6, 1 /93. Lemasson, J.-J., Fontenille, D., Lochaouarn, L., Dia, I., Simard, F., Ba, K., Diop, A., Diatta, M., Molez, J.F., 1997. Comparison of behavior and vector efciency of Anopheles gambiae and A. arabiensis (Diptera: Culucidae) in Barkedji, a Sahelian area of Senegal. J. Med. Entomol. 34, 396 /403. Lindsay, S.W., Wilkins, H.A., Zieler, H.A., Daly, R.J., Petrarca, V., Byass, P., 1991. Ability of Anopheles gambiae mosquitoes to transmit malaria during the dry and wet seasons in an area of irrigated rice cultivation in The Gambia. J. Trop. Med. Hyg. 94, 313 /324. Mukiama, T., Mwangi, R.W., 1990. Population and cytogenetic observations on Anopheles arabiensis Patton of Mwea irrigation scheme, Kenya. Insect Sci. Appl. 11, 119 /131. Mutero, C.M., Blank, H., Konradsen, F., van der Hoek, W., 2000. Water management for controlling the breeding of Anopheles mosquitoes in rice irrigation schemes in Kenya. Acta Trop. 76, 253 /263. Mwangi, R.W., Mukiama, T.K., 1992. Irrigation scheme or mosquito hazard; a case study in Mwea irrigation scheme. Hydrobiology 232, 19 /22. Pao Lingh Luh, 1984. The wet irrigation method of mosquito control in rice elds: An experience in intermittent irrigation