Joumal of the American Mosquito Control Association, 16(4\279-287, 2OOO Copyright O 2000 by the American Mosquito Control Association,Inc.

ANOPHELINE ECOLOGY AND MALARIA TRANSMISSION AT A NEW IRRIGATION PROJECT AREA (BARGI DAM) IN (CENTRAL )

NEERU SINGHlNo A. K. MISHRA

Malaria Research Centre (Field Station), Medical College Building, Jabalpur 4820O3, India

ABSTRACT, Anopheline ecology and malaria transmission were studied in a newly irrigated area of the Bargi Project, District Jabalpur, , Central India. Observations were made for 2 years (1993-95) in 10 villages along the Bargi irrigation canal, which are situated between 44 km (head end of canal) and 78 km (tail end of canal) from the dam site. Anopheles annularis was the predominant species in the head-end villages and its abundance was directly related to the opening of the canal, whereas Anopheles culicifacies was the most abundant species in tail-end villages, where irrigation is limited. Anopheles culicifacies showed 2 typical peaks not related to canal irrigation. Site-related differences in species prevalence were significant in both immatures and adults. Malaria infection was due to Plasmodium vivax and Plasmodiumfalciparum. The annual parasite incidence in children and adults was significantly higher in head-end villages (>4-fold) as compared to that in tail-end villages. However, seasonal trends in the prevalence of P. falciparum and P. vivax were the same in each group, with some fluctuations. In this study, preliminary results of the investigation are presented, demonstrating the trends in anopheline ecology and parasite prevalence in relation to the dynamics of irrigation development.

KEY WORDS Malaria, irrigation, Anopheles, Anopheles culicifocies, Plasmodium vivax, Plasmodium falci- DArum

INTRODUCTION Tava Project on a tributary of the Narmada in Hos- hangabad, Madhya Pradesh (Chaudhary 1996). Madhya Pradesh, Central India, the largest state This study was undertaken in villages along the in the country (442,841 km'), lies in the geograph- Bargi Canal. A major aspect of the study focused ical heart of India. It is a rural agricultural state on tracking the abundance and seasonality of marked by severe poverty and underdevelopment. anophelines and the incidence of malaria in a study The food grain production, infrastructure develop- area in the newly developed irigation project. This ment, and irrigation expansion in the state are the paper presents evidence of the formation of new lowest in the country (Tewari 1984). Construction disease foci in the hitherto less malarious area, pre- of a series of multipurpose projects for the devel- sumably under pressure of development. This case opment of the state have been proposed. The Nar- study may serve as a mode for rethinking malaria mada Valley Development Project is the largest of control progftlms under changing environmental the river valley projects so far proposed in India and ecological conditions due to developmental (Anonymous 1992). Balgi Dam, also known as projects. Rani Avanti Bai Sagar Dam, a multipurpose irri- gation and hydroelectric project (1974-88) is the MATERIALS AND METHODS lst completed project in Jabalpur. Water has been Study area: The study was carried out for 2 stored to its full capacity (422.26 m full river level) years (1993-95) in 10 villages (5 from Jabalpur and since 1990. The main left bank canal is 137 km 5 from Narshinghpur) along the Bargi main left long and total length of the canal along with dis- bank canal. Irrigation commenced in 1990 and the tributories is about 3,645 km (Anonymous 1984). irrigation system became fully operational in 1991. The total area proposed to be irrigated by the left The region falls within the lowland dry zone of the bank canal is 160,00O ha. Construction of a 194- state (elevation 360-411 m) with a generally flat km-long right bank canal to irrigate another terrain intemrpted by small foothills. Weather con- 250,000 ha of land has been proposed. Ttadition- ditions recorded at the nearest Agricultural Engi- ally, this study area was understood to be only neering College (Jabalpur) for the 3 years from mildly prone to malaria and hence no studies have 1993 to 1995 were, respectively, total annual rain- been done on mosquitoes and malaria in the dis- fall 1,486,2,083, and 1,417 mm; mean maximum tricts through which the left bank canal passes. temperature 31.7,3O.9O, and 31.60"C; mean mini- However, these developments might have resulted mum temperature 18.3, 18.2, and 18.8"C; and mean in ecological changes that can have a major impact relative humidity 59.90, 60.70, and 63.4OVo. on the incidence of mosquito-borne diseases. Such Head-end villages: Five villages, Dabhola, Bi- an effect has been observed during the develop- chua, Nunpur, Chhapara, and Bijori, are referred to mental projects of Sardar Sarovar, a downstream as head-end villages (total population, 2,302) be- project of Narmada in (Kalr:a 1992) and the cause they are located between 44 and 50 km (Fig.

279 280 Jounr*nl oF THE AMERTCANMoseurro CoNrRoL AssocrlrroN Vor. 16,No.4

lo D.id x t IlSs $gsskvitlm Scrle I @=5t& Aad od c ,!. Til6d a Mrir a0d B|'elasd MiDd Nrrui.riE 44 bercro Ro.dr Itisrb [email protected] i RdIrry lrc E I

{"c To Grd.rrn -d' faryr ed .io' To Itei Itittrtst N.ni4lt{t

Fig. l. Map showing Bargi left bank canal and location of study villages. l) from the dam site (the lst 39 km were not under soykans followed by wheat. The villages are under irrigation). Out of the 5,4have direct water supply routine dichlorodiphenyltrichloroethane (DDT) from the main canal tbrough minor canals (minors). spray (1 g*, 2 rounds each year) for malaria con- Only 1 village sharesits water from Chhapra minc. trol in the months of May and August. Dabhola is the only village through which the main MosQuito sampling: Indoor resting mosquitoes canal passes;the other villages are at about O.5-1.0 (per man hour) were collected onc@a month from km distant from the main canal. All the villages are November 1993 to October 1995 from 2 vrTlages, sparsely populated (scheduled caste 45Vo, sehed- I from the head-endvillages (24 collections) and 1 uled tribe SOVo,and general 5Vo). The total pro- from the tail-end villages (24 collections) using posed area under furigation is 649 ha. A very good standard entomological tecbriques (WHO 1975). network of subminors and distributories exists, yet Anopheles resting inside 4 s€lected sites located in at the time of this study, they were not cornpletely different parts of each village (2 human dwellings lined. Only 5OVoaf the lining work was completed and 2 cattle sheds) were sarnpled in the early morn- in patches- The water supply for irrigation is ex- ing (060O hrs) for 15 min each by a team of 2 insect cellent from October to April during each year. The collectors with flashlights and nlouth aspirators. AII averagedischarge flow in the main canal is 10-15 adult mosquitoes collected were identified morpho- m3/sec and in minor canals it is 1-2 m3lsec. The logically in the laboratory using standard keys main crops are ice, maize, and wheat. These vil- (Cbristophers 1933). lages were under regular hexachlorohexane (HCH) Bait collection' Anophelines attracted to hurnan spray (20O rflglm',3 rounds each year) for malaria and animal baits were sampled from dusk to dawn control in the months of May, July, and Septernber. at fixed stations once a month. Although we Use of HCH has been banned in India since 1997. planned to station human baits both indoors and Tail-end villages: Ponia, Kukla, Manakwara, Pi- outdoors, only I human bait was seatcd on stool in paria, and Dabkia were selected as tail-end villages the courtyard or verandah, where people sleep (total population, 3,896) because they are located throughout the year, because of limited number of at a distance of 75--78 km (Fig. l) from the dam staff for this study and unavailability ofvolunteers. site (of the l37-km-long main canal, only 80 km One additional collector caught anophelines attract- are constructed for irrigation). These villages are ed to buffalo baits (1 inside the house and 1 outside irrigated through 2 minors of the Hareri branch ca- the house) for 15 min each hour. nal (water flow 5-1O mTsec). The total proposed Only indoor resting anophelines were dissected area under irrigation is 868 ha, according to avail- for presence of sporozoites in salivary glands. Par- able records. Irrigation facilities and a network of ity determinations were done only in Anopheles irrigation canals exist, although water supply is not culicifocies Giles, because of its availability as good as in head-end villages. The main cro,p is throughout the year in both villages. Although DeceMsen 2000 Mnumn TRANsMIsstoN nt e New IRRtcATIoN PRoJEcr 281

Anopheles fluviatilis James, Anopheles annularis smears found positive for malaria parasites in a Van der Wulp, Anopheles subpictus Grassi, and An. year per 1,000 persons. culicifocies have been identified as species com- plexes and the members of each complex can be RESULTS distinguished by cytogenetic or DNA probe tech- niques, we did not attempt to identify the mosqui- Larval ecology toes further under the primitive field conditions. Larval sampling: Larval surveys were carried In all, 3,929 larvae were collected in 1,530 dips out once a month with standard dippers (250-ml in the head-end village and 3,451 larvae were col- capacity) from 1 village in each group. Larvae were lected in 1,526 dips in the tail-end village. Seven collected from the main canal and associated mi- anopheline species were identified after adult emer- nors and ditches and density per dip was recorded. gence from larvae collected from irrigation canals One dip every 2 m was taken in the slow-running and their various components. Percent emergence main canal and in the branch canal (15 dips in all). of the 5 most commonly collected species is shown About 1O-15 dips were taken from pools along the in Table l. In slow-flowing water of the main canal canal, ditches, and other breeding places depending and minors, An. culicifacies predominated and ac- on the area of the pools and pits. All these breeding counted for more than 8OVoof the anophelines col- sites were surveyed for 10 months in each year. In lected (range 7O-944o) in both villages. Larvae of minors, seepages, and rice fields dips were taken at anopheline species breed extensively in ditches and l-m intervals (10-15 dips) from 4 to 8 months in pools along the canals in both villages, whereas rice a year depending on the availability of water. Only fields support relatively less anopheline breeding 3rd- and 4th-stage larvae and pupae were retained and An. culicifocies was found breeding in freshly for adult emergence and identification. transplanted and fallow paddy fields. The other Malaria surveillance: Blood smears were pre- breeding places included slow-flowing water at the pared twice a month from all current fever cases grassy edges of agricultural fields and borrow pits. and people who reported fever during the preceding Anopheles subpictus was found mainly in rice 14 days (active case detection) from all the l0 vil- fields, ditches, and in pools along the canals in both lages. All people with fever were given 600 mg villages, although some larvae of this species were chloroquine as presumptive treatment. Blood found in the main canal and minors. Anopheles an- smears were stained with Giemsa and thick smears nularis was found primarily in seepage pools in the were examined under oil-immersion lens in the lab- head-end village, whereas few were recorded from oratory. Parasite-positive cases were given treat- the main canal and rice fields from either of the ment as per the National Anti Malaria Programme villages. Anopheles fluviatilis and, Anopheles ste- (NAMP). Under this regimen, people with Plas- phensi Liston were recorded from the head-end vil- modium vivax malaria were administered 600 mg lage only in small numbers, and were not recorded chloroquine as a single dose followed by 15 mg from the tail-end village at all. primaquine daily for 5 days and people with Plas- modium malaria were given 1,500 mg falciparum Indoor resting collections chloroquine on 3 days (6OO mg on day 0, 600 mg on day 1, and 300 mg on day 2) and 45 mg pri- In all, 8 anopheline species were collected in in- maquine as a single dose. Children were given pro- door resting collections from the head-end village portionately lower doses according to age. Pregnant and 9 were collected from the tail-end village. Data women and infants were not given primaquine. relating to the 3 most prevalent species collected Data analysis.' The following definitions apply are presented in Table 2. Anopheles annularis was to this study. Per man-hour density: the number of the most dominant species (45.5Vo) collected at the mosquitoes collected by I insect collector in t h. head-end village followed by An. culicifacies Parity rate: the number of parous mosquitoes in the (39.5%) and An. subpictus (l2.5Eo). At the tail-end total number of mosquitoes dissected. Slide positiv- village, An. culicifaci€r was the most prevalent spe- ity rate: total number of blood smears found posi- cies (59.9Vo) followed by An. subpictus (3l.lVo) tive for malaria parasite in the total number of and An. annularis (5.7Vo). Anopheles fluviatilis was blood smears examined. Slide falciparum rate: total less than l%o in both the groups of villages. The number of blood smears found positive for P. fal- percentage composition of these species differed ciparum in the total number of blood smears ex- significantly (X2 : 43.06, dt 2, P < 0.0001). amined. Plasmodium falciparum percentage (PfVo): Seasonal abundance of An. annularls at the head- total number of blood smears found positive for P. end village was directly related to the opening of falciparum in the total number of blood smears pos- the canals (October-April) during each year of the itive for the malaria parasite. Annual blood exam- study. Abundance of An. culicifocies was high ination rate (ABER): number of blood smeaf,s ex- throughout the year with a main peak in July-Sep- amined in a year in the total population under tember and a 2nd small peak in February-April in malaria surveillance. Annual parasite incidence both the villages. Anopheles subpictus was most (API): the proportion of the total number of blood abundant during the rainy season, that is, July-Sep- 282 JoURNAL oF THE AMERTcANMoseutro CoNrnol AssocrarroN Vol. 16,No.4

Table l. Anopheline larval density (per dip) and emergence in villages along Bargi Canal.

Eo emetgengel Larval An. An. Breeding sites density Total larvae culici- An. sub- An. fluvi An. surveyed (mean) (Vo) facies pictu.s annularis atilis stephensi Head end Main canal 2.8 7e (t9.4) 84.3 2.2 12.O 0.o 0.0 Minors 1.0 180(4.6) 94.2 3.8 1.9 0.o 0.0 Pools along canal z.J 68r (17.3) 76.2 19.8 J.Z 0.0 0.8 Ditches A1 1168(29.7) 67.4 25.8 A< 0.0 1.1 Pools due to seepage 2.3 lsl (3.8) 54.O 0.0 46.O 0.0 0.0 Rice fields 3.0 360 (9.2) 44.2 43.5 to.2 z.o 0.0 Others 2.1 62s (t5.9) 40.1 23.O 34.4 1.3 0.6 All 2.53 3929 65.8 16.9 16.03 0.s 0.36 Tail end Main canal (branch canal) )4 660(19.1) 70.4 l7.o 12.O 0.0 0.0 Minors 1.5 437 (12.6) 79.O 19.0 0.9 0.0 0.0 Pools along canal z.J 679(19.7) 73.8 2t.5 4.6 0.0 0.0 Ditches J.Z 1306(37.8) 52.6 44.3 1.5 0.0 0.0 Pools due to seepage 1.0 ee (2.8) 67.O 33.0 0.0 0.0 0.0 Rice fields 1.9 22O(6.4) 35.3 59.8 4.9 0.0 0.0 Others 0.8 50 ( 1.4) 61.5 38.5 0.0 0.0 0.0 All 1.87 3451 62.8 JJ.J -t-+ | 0.0 0.o

I An., Anopheles. tember in 2 villages. Seasonal fluctuations were less villages, respectively. Anopheles culicifacies was predictable for An. fluviatilis because of their small most prevalent (1.1 per bait night) followed by An. numbers. Other species such as Anopheles theobal- annularis (0.33 per bait night) and An. subpictus di Giles, Anopheles splendidus Koidzumi, and (0.2 per bait night) at the head-end village. From Anopheles vagus Donitz were only rarely collected. the tail-end village, An. culicifacies was the most Cattle bait collections showed the same trend of common (0.6 per bait night) followed by An. sub- Anopheles prevalence in 2 villages (Table 3) as was pictus (O.3 per bait night) and An. annularis (O.l found in the indoor restins collections. per bait night).

Human bait collection Salivary gland dissection

Only 19 and 1l females were collected during Overall, 4OOAn. culicifacies, 13 An. fluviatilis, 12 human bait collections at head-end and tail-end 5O An. annularis, and 50 An. subpictus were dis-

Table 2. Number of anopheline adults collected during indoor resting catches (man-hour density, 1993-95) in villages along Bargi Canal.l

Head-end village Tail-end villase

An. An. An. An. An. An. Month culicifacies annularis subpictus Total2 culicifacies annularis subpictus Total2

Nov. l2 (10.1) 9t ('76.s) 3 (2.s) 119 re (67.9) 0 (0.0) 6 (21.4) 28 Dec. 17(r7.3) 65(66.3) 3(3. r) 98 r7 (43.6) 4 (r0.3) ls (38.s) 39 Jan. 22(23.9) s9(64.1) 8 (8.7) 92 2s(1o.2) 3 (6.r) 2r (42.9) 49 Feb. 47 (27.6) r2o (70.6) 2 (r.2) t70 37 (s9;7) 4 (6.5) 20 (32.3) 62 March 73(sz.r) 64 (45.7) 2 (r.4) 140 49(6r.3) l2 (15.0) l9 (23.8) 80 ApriI 30 (32.6) 60 (6s.2) )t))\ 92 so (79.4) 0 (0.0) t3 (20.6) 63 May 20 (37.O) 29(s3.7) s (e.3)

I An. Anopheles. Numbers in pilentheses indicate percentage of total collected during month for each village. 2 Anopheles fluviatilis, An. theobaldi, An. splendidus, An. stephensi, and An. vagus are included in the head-end village and these species along with An. nigenimus are included in tail-end village. DscpMssn 2000 MeLnml TRANsMIssIoN AT A NEw IRRIcATIoN PRoJEcr

Table 3. Number of anophelines collected on animal baits in Bargi Canal villages (1993-94).1

Head-end villaee Tail-end village

An. An. An. An. An. An. Month culicifacies annularis subpictus culicifacies annularis subpictus

Nov. 2 68 J 74 Dec. 5 97 3 105 7 I i; Jan. 8 80 2 92 z 0 23 25 Feb. 43 52 I ro4 t7 o t4 32 March 35 48 2 86 38 I 645 April 8 z-) 7 38 7 2 15 24 May 6 16 ll 33 9 0 12 21 June I 8 5 t4 2 0 57 July 13 2 6 2l 5 I II JJ Aug. 180 IJ 30 224 153 I 13 r72 Sept. 53 0 2l 75 4l 0 22 64 Oct. 7 44 2 53 IJ 3 580 Total 308 (12.8) 4sl (18.8) 72 (3.O) 844(3s.3) 366 (16.6) l2 (0.5) ls3 (6.9) s36(24.3)

tAn., Anopheles. Figures in parentheses indicate per bait per night. sected from head-end villages to determine sporo- cases began to appear from April to May and con- zoite infection. Only I An. culicifucies was positive tinued until September to October. Plasmodiumfal- for sporozoites in the month of August 1995. None ciparum cases were found mainly from July to De- of the remaining species dissected was sporozoite cember and very few P. falciparum cases were positive. From the tail-end village, the salivary recorded during March to June. Overall, PfVo was glands from 136 An. culicifacies, 25 An. annularis, higher in a&lJts (69qo) as compared to children and 25 An. subpictur were examined for sporozo- (39Vo) at head-end villages (Table 5), which is high- ites. None was positive for sporozoites. ly significant statistically (P < 0.0001), whereas at tail-end villages (Table 6) the difference inPf%o be- tween adults (66.5Vo) and children (53.4Vo) was sig- Parity status nificant at a low level (P < O.O5). In contrast, P. Parity rates of An. culicifacies varied from 56 to falciparum g.rmetocyte carriers were significantly 69Vo (95Vo, CI 4.3-3O.3Vo) at the head-end village higher (P < 0.001) in children (l6.5Vo) as com- and 4l-75%o (95%, CI ll-57Vo) at the tail-end vil- pared to adtrlts (6.7Eo) at head-end villages, where- lage (Table 4). The difference between the head- as at tail-end villages the difference in P. falcipa- end village and the tail-end village is significant rum gametocytes carriers was not significant statistically (X' : 2O.O6,dt 9, P < 0.01). between children (l6Vo) and adults (107a). When the data were pooled over villages and years and the ABER and API were compared be- Pattern of malaria prevalence in active tween children and adults. both ABER and API surveillance were significantly higher in children (ABER, 76.7; Data were combined per month over villages API, 216) and adults (ABER, 73; API, 140) at (Tables 5 and 6) to examine seasonal changes in head-end villages (P < 0.00001) as compared to malaria prevalence between children (<14 years children (ABER, 27; APl,57) and adults (ABER, old) and adults (>14 years old). A similar seasonal 26; API, 33) at tail-end villages (P < 0.00001). pattern of P. vivax and P. falciparzrn infections was Overall, 2.4 times more malaria cases and 2.2 times found in both children and adults in each year of more P. falciparum cases were recorded from head- the study with some fluctuations. Plasmodium vivax end villages than from tail-end villages. Further,

Table 4. Parity status of indoor-restingAnopheles culicifacies caught in villages along the Bargi Canal (Nov. 1993- Oct. 1995).

Head-end village Tail-end village Total Nulli- Parity Total Nulli- Parity dissected parous (vo) dissected parous (Vo) Nov.-Jan. 24 14 lO s8.3 20 12 8 60.0 Feb.-April 55 31 24 56.4 37 l9 18 51.3 May-July 44 28 16 63.6 229 13 40.9 Aug.-Oct. 64 44 20 68.8 7l 53 18 74.6 Total 187 117 70 62.6 t50 93 57 62.O 284 Jouruul oF THE AMERICIN Mosquno Colrnol AssoctettoN Vor. 16.No.4

Table 5. Malaria prevalencein children (< 14 years old) and adults (> 14 years old) at head-endvillages along Bargi Canal (1993-95).1

Month/year BSE Pf SPR BSE Pv Pf SPR PfVo Nov. 1993 l19 t7 30 39.5 63.8 455 0 67 14.7 100.0 r994 34 I 6 20.6 85.7 80 0 15 18.8 100.0 Dec. 1993 43 2 \2 32.6 85.7 t)t 0 22 t4.o 100.0 1994 27 5 5 37.O 50.0 83 0 t2 t4.s 100.0 Jan. 1994 4l I 5 14.6 83.3 94 2 16.0 86.7 r995 3l 2 J 16.1 60.0 44 z 6.8 33.3 Feb. 1994 27 I -) 14.8 75.O 0 4 7.8 100.0 1995 10 I I 20.o 50.0 2 0 r 1.8 0.0 March 1994 J+ 2 0 5.9 0.0 80 I 3 5.0 75.O 1995 4 2 0 50.0 0.0 8 1 0 12.5 0.0 April 7994 )U 9 2 22.O 18.2 89 6 l0 18.0 62.5 1995 56 7 0 12.5 o.0 55 ll 0 20.o 0.0 Muy 1994 JZ 9 I 31.3 10.0 69 6 9 21.7 60.0 1995 4l 8 0 19.5 0.0 69 l9 I 29.O 5.0 June 1994 +J 7 I 18.6 t2.5 t07 5 3 7.5 37.5 r995 39 5 0 12.8 0.0 44 8 0 t8.2 0.0 JuIy 1994 89 74 7 23.6 33.3 2t3 4 8 5.6 66.7 1995 62 ll 0 17.7 0.0 64 9 1l 3l .3 55.0 August r994 80 23 16 48.8 41.0 t26 8 25.4 75.0 1995 67 7 8 22.4 53.3 76 8 6 18.4 42.9 Sept. t994 tl7 3l 13 37.6 29.5 211 A 57 28.9 93.4 1995 55 6 s 20.o 45.5 84 J l4 20.2 82.4 Oct. 1994 87 22 l5 42.5 40.5 158 0 3'l zJ.+ 100.0 1995 JJ I I 6.1 50.0 63 3 l0 20.6 76.9 ' BSE, blood slide examined; W, Plasmodium vivul Pf, P. falciparum; SPR, slide positivity fttet PfEo, P. .falciparum percentage

API among infants at head-end villages was 161 the District since 1991 (Singh et al. 1999a). and 22, respectively, during the lst and 2nd years, This malaria is not responsive to control measures. whereas at tail-end villages the API among infants Thus, malaria control requires specific approaches was 0 during both years (data not shown). and control strategies for irrigation projects, which emphasizes the need for better understanding of the epidemiologic features and transmission dynamics DISCUSSION of malaria. The Bargi irrigation project presented Large-scale irrigation developments often have an oportunity for a case study on mosquito ecology resulted in increased human malaria incidence and malaria in relation to irrigation development. (Amerasinghe and Indrajith 1994, Tyagi and Unforhrnately, no figures are available to show how Chaudhary 1997). The effect of irrigation projects many anopheline species or how much malaria ex- on the incidence of malaria varies with the type of isted in the study area before the development of irrigation employed, with the physiographic and the irrigation canal. climatic features of the area irrigated, and with the Analysis of the results of the salivary gland dis- methods of cultivation practiced therein (Singh and sections suggested that the only malaria vector is Puri 1951). Mandla, Jabalpur, and Seoni districts An. culicifacies, as recorded earlier (Singh et al. are affected by the Bargi irrigation project. A major 1996, 1999b). However, too few mosquitoes were epidemic occurred in many villages near the Bargi dissected to establish this definitively. A larger Dam reservoir (Singh et al. 1997) and a steady in- number of salivary glands could not be dissected crease in malaria incidence has been recorded in because of the lone distance between field sites and DpcsN4ssn2000 Mnr-nnrn TneNsprtssroNlr n Nr,w InnrcnnoN Pnorscr 285

Table 6. Malaria prevalence in children (< 14 years old) and adults (> 14 years old) at tail-end villages along Bargi Canal (1993-95).1

Children

Month/year BSE Pv Pf SPR Pf SPR PfVa Nov. 1993 63 5 22 42.9 81.5 r79 4 23 15.l 85.2 t994 l5 0 I 26.7 100.0 3l o 5 l6.l 100.0 Dec. r993 3l I 8 29.O 88.9 720 u 15.3 100.0 r994 18 1 I It.l 50.0 420 z 4.8 100.0 Jan. 1994 20 2 4 30.0 66.7 620 0 0.0 0.0 t995 l6 I 2 18.8 66.7 36 l 2 8.3 66.7 Feb. 1994 t9 0 4 21.1 100.0 520 z 3.8 100.0 1995 JZ 3 2 15.6 40.0 450 0 0.0 0.0 March 1994 22 2 I t3.6 JJ.J 340 4 r 1.8 100.0 1995 2l 0 0 0.0 0.0 502 1 6.0 JJ.J April 1994 16 I 0 6.3 0.0 250 2 8.0 100.0 t995 22 0 0 0.0 0.0 40 1 0 2.5 0.0 May 1994 26 4 0 15.4 0.0 232 0 8.'7 0.0 1995 29 -) 2 t7.2 40.0 61 5 I 9.8 16.7 June 1994 29 -) I 13.8 25.O 21 4 0 19.0 0.0 1995 31 7 0 22.6 0.0 z)+ o 17.4 0.0 July 1994 l6 I I 12.5 50.0 480 J 6.3 100.0 1995 47 J 0 6.4 0.0 728 0 ll.1 0.0 Aug. t994 28 5 J 28.6 37.5 552 6 14.5 75.0 r995 36 z I 8.3 JJ.J 670 5 4.5 100.0 Sept. 'l r994 44 7 31.8 50.0 101 l0 20 29.7 66.7 1995 44 5 J 18.2 37.5 852 8 I1.8 80.0 Oct. r994 47 2 T4 34.O 87.5 826 22 34.1 78.6 t995 l8 0 2 ll.l 100.0 432 l0 27.9 83.3 I BSE, blood slide examined; Pv, Plasmodiumvivu:Pf, P. falciparuml SPR, slide positivity rateiPfTo, P. falciparum percentage.

the laboratory (25 km from the head-end village numbers during a limited period of the year in lon- and 50 km from the tail-end village). As a result, gitudinal studies carried out simultaneously in con- many mosquitoes died in transit (a constraint im- nection with another research project in nonirrigat- posed by the nonavailability of enzyme-linked im- ed villages about 20 km from the study area (Singh, munosorbent assay technology). Other known vec- unpublished data) or in other nonirrigated areas of tors, An. fluviatilis (Kulkarni 1987, Subbarao et al. Jabalpur (Singh and Sharma 1989, Singh and Mish- 1992) and An. stephensi (Sharma, 1996), are pres- ra 1997) and in Mandla (Singh et al. 1996). The ent in the study area in small numbers from head- larval abundance pattern generally agreed with the end villages. These 2 species may play a major role adult pattern, but this was difficult to quantify be- in malaria transmission if environmental changes cause of differences in habitat size and the clumped favor increased breeding under conditions of canal distribution of larvae within habitats. Interestingly, irrigation. all these species were highly zoophagic, as indicat- In Sri Lanka, in the Mahaveli Project after irri- ed by comparative human versus animal bait biting gation development, an increase in abundance of collections in the study area. However, An. subpic- An. annularis and An. subpictus was reported trr,r was incriminated as the vector in Baster, Mad- (Amerasinghe and Ariyasena 1990, Amerasinghe et hya Pradesh (Kulkarni 1983) and An. annularis was al. 1991). Their data are consistent with high An. incriminated in Koraput, Orissa (Gunasekaran et al. annularis and.An. subpictus catches recorded from 1989). Subsequent stabilization of the irrigation ca- head-end and tail-end villages, respectively in this nals and an increase in natural and planted vege- study. The 2 species were found in insigniflcant tation is likely to encourage the establishment of 286 JounNlL oF THE AMERTCaNMoseurro CoNTRoL Assoctelox VoL. 16,No.4 some ofthese anopheline species as significant vec- care program based on local transmission involving tors of malaria, as recorded elsewhere (Tyagi and multisectoral action and community participation to Chaudhary 1997). prevent the spread of disease in the whole region. Perennial irrigation systems tend toward overir- rigation. The problem is worsened by the fact that canals are not lined and no drainage system is ACKNOWLEDGMENTS available. is the largest monsoon Grateful thanks are due to M. H. Birley (Liver- river of Madhya Pradesh, hence the bulk of the run- pool School of Tropical Medicine, Liverpool, Unit- off water is generated during monsoon months that ed Kingdom) for reviewing the manuscript. The are marked by spells of intense and heavy rainfall. technical assistance rendered by the Field Staff of Flooding of tracts by the overflow of silt-laden wa- the Malaria Research Centre, Field Station Jabalpur, ter from the Narmada River during the monsoon is gratefully acknowledged. season creates scours in the bed. The original gra- dient was altered because of silt deposits, resulting in the obstruction of the free flow of water. REF'ERENCES CITED At head-end villages, temporary obstructions are Amerasinghe FP, Ariyasena TG. 1990. Larval survey of commonly created by villagers during slow release surface water breeding mosquitoes during irrigation de- to maximize the amount of water for their crops. velopment in the Mahaweli Project, Si Lanka. J Med During peak release, water escapes its banks and Entomol 27:789-8O2. spreads to low-lying areas. In tail-end villages, non- Amerasinghe FP, Indrajith NG. 1994. Post-irrigation breeding patterns of surface water mosquitoes in the availability of sufficient water for the crop is a com- Mahaweli Project, Sri Lanka and comparisons with pre- phenomenon. mon Here, the illicit removal of water ceding developmental phases. J Med Entomol 3:517- is a problem. Needless to say, the irrigation canal 520. has provided additional breeding sites for anophe- Amerasinghe FP, Priyanie H, Amerasinghe JS, Malik P, lines by slowing the current in many places in Wirtz RA. 1991. Anopheline ecology and malaria in- which pools form. Intense breeding was recorded fection during the irrigation development of an area of in temporary pools in the irrigation canals imme- the Mahaweli Project, Sri Lanka. Am J Trop Med Hyg diately after the water was shut off, as frequently 45:226-235. occurs when repairs of the canal bank are neces- Anonymous. 1984. Bargi Multipurpose Project and Bargi Diversion Scheme Jabalpur, Madhya Pradesh, India: sary. Moreover, during the nonirrigation season, Bargi Multipurpose Project Authority. 11 p. rainwater pools that support heavy breeding of Anonymous. 1992. Narmada Valley Development Project anophelines are formed in the canals. Bhopal, Madhya Pradesh, India: Narmada Valley De- The incidence of malaria was very high in the velopment Authority. 12 p. study villages, particularly in the head-end villages Birley MH. 1989. Guidelines fon forecasting the vector- in spite of the low human biting and sporozoite rate borne disease implications of water resources devel- in An. culicifacies. Despite being a zoophilic spe- opment PEEM Guideline Series 2. WHO/VBC/89.6. cies, An. culicifacies is known to be capable of Geneva: World Health Organization. transmitting malaria because of its dispersive be- Birley MH. 1991. Guidelines for forecasting the vector- borne disease implications of water resources devel- havior and longevity (Curtis and Rawlings 1980, opment 2nd ed. WHO/CWS/91.3. Geneva: World Rawlings et al. 1981). Thus, the incidence of ma- Health Organization. laria in the villages studied here apparently is di- Chaudhary N. 1996. Bandh aur Beemari lTava Dam, a rectly related to irrigation. In tail-end villages, studyf ln Disha samvad Hoshangabad, Madhya Pra- where less irrigation is present, villages are less ma- desh, India: Gram Seva Samiti. p l-41. larious. The decline in ABER and API during 2nd Christophers SR. 1933. The.fauna of British India includ- year of the study may be related to surveillance and ing Ceylon and Burma Vol 4. London: Taylor and Fran- drug distribution by project staff. cis. Curtis CE Rawlings P 1980. A preliminary study of dis- Of concern is that irrigation-related malaria is persal and survival of Anopheles culicifacies in relation probably more prevalent in Madhya Pradesh now to the possibility of inhibiting the spread of insecticide than ever before because of the construction of var- resistance. Ecol Entomol 5: I l-17. ious major and minor developmental projects. Gunasekaran K, Sahu SS, Parida SK, Sadanandane C, Health issues seem to have been neglected in many Jambulingam B Das PK. 1989. Anopheline fauna of development projects (Birley 1991). Several guide- Koraput District, Orissa State with particular reference lines have been published to facilitate the assess- to transmission of malaria. Indian J Med Re,g 89:34O- ment of health impacts for water resource projects 343. and other types of development (Birley 1989). Kalra NL. 1992. Status report on malaria and other health related aspects of the (SSP) projects and recommen- However, little has been done in practice. Poor en- dations regarding short-term and long-term remedial gineering design is difficult to correct after con- measures New Delhi, India: World Bank Supervisory struction, and hence early planning is critical. It is Mission Report. already too late for Bargi to prevent some of these Kulkarni SM. 1983. Detection of sporozoites in Anopheles consequences. Construction is progressing and ac- subpictus in Bastar District, Madhya Pradesh. Indian J tion is required now to develop an effective health Malariol 20:159-160. DecBNasen2000 MALARTATRANsMrssroN lr n Ntw IRRIGATIONPRoJEcr 287

Kulkarni SM. 1987. Feeding behaviour of anopheline Singh N, Shukla MM, Chand SK, Sharma VP 1997. Out- mosquitoes in an area endemic for malaria in Bastar break of falciparum malaria in submerged villages of District, Madhya Pradesh. Indian J Malariol 24:163- Narayanganj PHC, district Mandla due to Narmada Ir- 171. rigation Project, central India (Madhya Pradesh). Carr Rawlings B Curtis CE, Wickremasinghe MB, Lines J. Sci 73:686-691. 198 1. The influence of age and season on dispersal and Singh N, Singh OP, Sharma VP. 1996. Dynamics of ma- recapture of Anopheles culicifacies in Sri Lanka. Ecol laria transmission in forested and deforested region of Entomol 6:3O7-319. , central India, Madhya Pradesh. An J Sharma VP 1996. Re-emergence of malaria in India. /n- Mosq Control Assoc l2:ll5-124. dian I Med Res LO3:26-45. Subbarao SK, Vasantha K, Joshi H, Raghvendra K, Usha Singh J, Puri IM. 1951. Malaria incidental to engineering devi C, Sathyanarayan TS, Cochrane AH, Nussenzweig construction. Health Bull 32:7J4. RS, Sharma VP. 1992. Role of Anopheles culicifades Singh N, Mehra RK, Sharma VP 1999a. Malaria and the sibling species in malaria transmission in Madhya Pra- Narmada River development, in India: a case study of desh, India. Trans R Soc Trop Med Hyg 86:613-614. Bargi Dam. Ann Trop Med Parasitol 93:.477-488. Tewari DN. 1984. Primitive tribes of Madhya Pradesh. Singh N, Mishra AK. 1997. Effi,cacy of light traps in sam- Strategy for development New Delhi, India: Govern- pling malaria vectors in different ecological zones in ment of India, Ministry of Home affairs, Tribal Devel- central India. Southeast Asian J Trop Med Public opment Division. 79 p. Health 28:196-2O2. Tyagi BK, Chaudhary RC. 1997. Outbreak of falciparum Singh N, Mishra AK, Chand SK, Sharma VP 1999b. Pop- malaria in the Thar Desert (India), with particular em- ulation dynamics of Anopheles culicifacies and malaria phasis on physiographic changes brought about by ex- in Tribal area of Central lndia. J Am Mosq Control As- tensive canalization and their impact on vector density soc 15:283-29O. and dissemination. J Arid Environ 36:541-555. Singh N, Sharma VP 1989. Persistent malaria transmis- WHO [World Health Organizattonf. 1975. Manual on sion in Kundam Block, District Jabalpur (M.P.). Indian practical entomology in malaria part II Geneva: World J Malariol 26:.11. Health Orsanization.