Establishment, Spread and Impact of Neochetina spp.
on Water Hyacinth in Lake Victoria, Kenya
G.S. Ochiel*, S.W. Njoka*, A.M. Mailu† and W. Gitonga‡
Abstract
The Kenya Agricultural Research Institute imported 12,300 curculionid weevils (Neochetina spp.) from
diverse sources, for biological control of water hyacinth in Lake Victoria, as part of the World Bank-funded
Lake Victoria Environmental Management Project in East Africa. In addition to the rearing and quarantine
facility at Muguga, a second rearing facility was established in 1996 at Kibos, near Lake Victoria. The Kibos
rearing facility and two community rearing facilities at the lakeshores, have produced approximately 100,000
tami chynn
adult weevils and 42,000 weevil eggs over a three-year period. Since January 1997, some 73,500 Neochetina
weevils have been released at 29 sites and an additional 10,000 redistributed at several sites. Visual
observations and regular sampling monitored the establishment and spread and also evaluated the impact of
Neochetina weevils on water hyacinth. Within two years, weevils were established at 55% of release sites and
were being recovered 50 km from release sites. Post-release sampling data from four release sites in Berkeley,
Kisumu and Kendu bays, indicated a reduction in leaf length, laminar area and fresh weight of water hyacinth,
and a significant increase in number of weevil feeding scars and adult weevils per square metre. Three years
after the initial weevil releases, the combined mean number of weevils per plant for Kisumu, Nyakach, Kendu
and Homa bays, was estimated to be six, well above the critical threshold of five weevils per plant. N. bruchi
was the dominant species accounting for 73.3% of the total weevil population. Thus, under Lake Victoria
conditions, the critical threshold was attained within 2–3 years of the initial releases.
L AKE Victoria (area ca 69,000 km2), shared by the three East African countries, Kenya (6%), Uganda (43%) and Tanzania (51%), is the world’s second-largest freshwater lake (Figure 1). In 1989, it was invaded by water hyacinth and its presence in the Kenyan part was confirmed in 1992. The origin of the infestation is presumed to be in the River Kagera Basin in Rwanda. At peak infestation in 1997, the area covered by the weed in East Africa was more than 15,000 ha. The tropical aquatic weed of South Amer-ican origin, has adverse impacts on the health, energy, water and transport sectors (Harley 1990; Harley et al. 1996). The weed presented an enormous challenge for biological control in East Africa.
As early as 1993, the Kenya Agricultural Research Institute (KARI) imported water hyacinth weevils, Neochetina bruchi and N. eichhorniae, from the Plant Health Management Division of the International Institute for Tropical Agriculture in Benin. These wee-vils, considered the most important biological control agents against the water hyacinth, have had notable success outside East Africa (Harley 1990; Julien and Griffiths 1998; Julien et al. 1999). However, host-spe-
*Kenya Agricultural Research Institute, National Fibre Research Centre Kibos, PO Box 1490, Kisumu, Kenya.
Email:ochiel@swiftkisumu,
njoka@swiftkisumu
†Kenya Agricultural Research Institute, PO Box 57811, Nairobi, Kenya. Email: odalis@
‡Kenya Agricultural Research Institute, National Agricultural Research Centre, Muguga, PO Box 30148,
Nairobi, Kenya. Email: ke
89
cificity tests were ordered in Kenya and Uganda, before releases in Lake Victoria were allowed. Neochetina weevils were released in Lake Kyoga, Uganda in 1993 (Ogwang and Molo 1997, 1999) and in Lake Victoria in 1996 (James Ogwang, pers. comm.). The first weevil releases in Kenya were in Lake Naivasha, which had water hyacinth since the mid 1980s (Aggrey Mambiri, pers. comm.). Neo-chetina weevils were released in the Kenyan part of Lake Victoria in 1997 (Ochiel et al. 1999; Mailu et al. 1999), while in Tanzania, Mallya (1999) reported the releases of Neochetina weevils in the Pangani and Sigi rivers in 1995, and in Lake Victoria in 1996.
This paper presents recent results from a program of classical biological control against water hyacinth in Lake Victoria, implemented by KARI under the Lake Victoria Environmental Management Project.
Materials and Methods
Mass rearing and releases of Neochetina spp. Since 1996, Kenya Plant Health and Inspectorate Services has allowed KARI to import adult N. bruch i and N. eichhorniae from Uganda, South Africa and Australia for the biological control of water hyacinth in Lake Victoria. KARI established a second weevil rearing facility in December 1996, at the National Fibre Research Centre (NFRC), Kibos, near Lake Victoria. ‘Breeding stock’ for the Kibos rearing facility was obtained from the quarantined mass-re
aring facility at the National Agricultural Research Centre, Muguga, near Nairobi. The breeding material consisted of mature adult Neochetina weevils and host plants inoculated with weevil eggs. Later, adult
Figure 1.Lake Victoria and surrounds
90
Neochetina weevils were imported from Uganda for mass rearing. Julien et al. (1999) describe in detail rearing and harvesting techniques for Neochetina weevils from plastic tubs, rearing pools and galva-nised corrugated iron sheet tanks, all of which have been in use at the Kibos rearing facility. Addition-ally, ‘Technotank’ PVC tanks (120 × 60 cm; 230 L), with sawn-off lids, have been used to rear the weevils at Sango Rota and Nyamware beaches and at Ogenya Primary School (community-based rearing facilities near the lake). Fertiliser NPK 17:17:17 and dried cow-dung were added to the rearing containers once a month to maintain plant vigour.
Weevils were harvested for field releases as described by Julien et al. (1999). Neochetina weevils imported from South Africa were released in Lake Victoria in 1997 and further releases were carried out with weevils reared at NFRC Kibos and community rearing facilities. Hyacinth plants infested with weevil life stages and adult weevils were used for releases. Adult weevils were fed on fresh leaves and petioles in plastic jars before transporting them to release sites. Release techniques included planting host plants infested with weevil life stages among hyacinth plants and tipping adult weevils from the pl
astic containers onto hyacinth plants. Weevils were also released at sites more than 50 m from the shoreline. Canoes were used to release at sites that were inaccessible by motor vehicle or on foot.
Monitoring the establishment and spread of Neochetina weevils
We recorded petiole damage by weevil larvae, fresh adult feeding scars and the number of adult weevils on water hyacinth at release sites with resi-dent mats of water hyacinth and at non-release sites.These visible signs are indicators of an establishing or established weevil population at a given site. Weevil recovery at non-release sites indicated weevil spread on water hyacinth.
Evaluation of the impact of Neochetina spp. weevils on water hyacinth
Using a modified sampling protocol developed at the Commonwealth Scientific and Industrial Research Organization (CSIRO), Australia, we eval-uated the impact of Neochetina spp. on water hya-cinth. The objectives of the sampling were to: (1) evaluate water hyacinth growth parameters; (2) quantify weevil feeding damage; and (3) estimate weevil populations. A half-metre2 quadrat was thrown randomly on mats of hyacinth plants. The number of plants per quadrat was recorded. For each of 10 or 30 plants from the quadrat or nearby, the fol-lowing parameters were recorded: fresh weight; leaf laminar area; leaf length; number of feeding scars; number of weevils per plant; and number of ad
ult weevils per square metre (mean number of weevils per plant × number of plants per quadrat). Rapid assessment of weevil populations was done by counting the number of weevils from each of 10 or 20 randomly selected plants at selected sites.
Results
Importation of Neochetina weevils
Between 1996 and 1998, KARI imported 12,300 Neochetina weevils from Australia, South Africa and Uganda, for mass rearing and releases on water hya-cinth in Lake Victoria (Table 1).
Table 1.Importations into Kenya of Neochetina weevils for biological control of water hyacinth in Lake Victoria Species Year
imported Number
Purpose Source Neochetina bruchi 19961300Mass rearing Uganda
1997
2000Mass
rearing/releases
Australia
1998 1000a a Batch did not survive Releases South Africa
Neochetina eichhorniae19975000Mass rearing/releases South Africa 19972000Mass rearing/releases Australia
19981000Releases South Africa Total12,300
91
From December 1996 to December 1999, the Kibos rearing facility and community rearing facilities pro-duced approximately 100,000 adult weevils, of which 25,000 were for ‘breeding stock’ and for releases in Lake Naivasha and Nairobi Dam.
Between January 1997 and December 1999, approximately 73,200 adult weevils were released at 29 sites in Kisumu, Nyando, Rachuonyo, Bondo, Homa Bay, Migori, Suba and Busia districts (Table 2). An additional 10,000 weevils were redistributed from the Homa Bay Pier and Police Pier release sites to other sites within the Kisumu District. Monitoring the establishment and spread of Neochetina weevils
Monitoring in January 1999 confirmed that the weevils were firmly established at 16 sites in 7 dis-tricts along the Lake Victoria shoreline (equivalent to 55% of the sites). Weevil recoveries were also made at distances ranging from 5–50 km from the nearest release sites.
Evaluation of the impact of Neochetina spp. weevils on water hyacinth
In general, post-release sampling data collected (November 1997 to May 1998) at four selected release sites in Berkeley, Kisumu and Kendu Bays, indicated a suppression of plant growth parameters (fresh weight, leaf laminar area and leaf length) and substan-tial increases in number of feeding scars and adult weevils per plant (Table 3). Fresh weight reduction was noted at a single site, Bukoma Beach. Leaf length reduction was noted at two sites, while leaf laminar area reduction was evident at Sio Port and Bukoma. The number of feeding scars and adult weevils per plant increased at all sites.
Estimations of weevil populations
Post-release sampling of water hyacinth at six selected sites in three bays (May–December 1999), gave a combined mean number of 6.0 Neochetina weevils per plant, with actual number of weevils per plant ranging from 0 to 32 (Table 4). Table 4 also shows that N. bruchi was the dominant of the two we
evil species, accounting for 73.3% of the total weevil population.
Discussion
Importation of additional biological control agents, the moth Niphograpta albiguttalis, the mite Orthoga-lumna terebrantis and the hemipteran bug Eccrito-tarsus catarinensis, to augment biological control efforts by Neochetina weevils, is recommended.
Rearing pools, which are easier to manage and have a larger capacity, are preferred over both plastic basins and tanks and galvanised iron sheet tanks. Tub rearing was found to be labour-intensive and time-consuming.
Tubs may, however, be used for ‘demonstration mass rearing units’ in schools and community-based rearing facilities near the lake.
Releases on floating mats assisted in the redistribu-tion and spread to non-release sites. Wind and water currents were responsible for the spread of weevils on floating mats of water hyacinth. Under the environ-mental conditions of Lake Victoria, weevils estab-lished quite rapidly.
At a regional level, monitoring the water hyacinth infestation pattern using aerial reconnaissance photo
g-raphy, ground truthing and satellite imagery has been proposed. At a national level, monitoring and evalua-tion of the impact of weevils on water hyacinth, redis-tribution to areas with low weevil populations and scouting for new infestations should continue.
Weevil damage has been held primarily responsible for the reduction of the weed cover by up to 80%, from the peak infestation of 6000 ha in 1998 (Synoptics, Integrated Remote Sensing and GIS Applications, The Netherlands). By late 1999, water hyacinth in the Kenyan part of the lake was no longer capable of flow-ering and producing ramets (daughter plants). This has been attributed to weevil damage and opportunistic fungi. The El Niño flooding of 1997 may have physi-cally destroyed plants by washing them ashore.
Ecological succession of water hyacinth by emer-gent plant species, mainly papyrus (Cyperus papyrus) and hippograss (V ossia cuspidata), has been noted (Ochiel, personal observations). This phenomenon has also been observed in Lake Kyoga, Uganda, following the successful biological control of water hyacinth by Neochetina weevils. However, this is short-lived and the secondary vegetation will disappear after the degraded hyacinth substratum supporting it eventually sinks.
The long-term approach to water hyacinth manage-ment and indeed other floating or submerged aqua
tic weeds, should focus on curbing the discharge of efflu-ents into Lake Victoria from surrounding urban settle-ments, agricultural and industrial activities.
92
Table 2. Releases of Neochetina at sites in Lake Victoria, Kenya, January 1997 to September 1999.
Site–Grid Reference Release dates Life stage
Eggs Adults
Police Pier 0°5.5'S;34°44.3'E 23.1.97–18.2.98 13 850 5 553
Fisheries Pond 0°5.4'S;34°44.0'E 23.1.97–15.4.97 3 680 1 000
Golf Club 0°5.4'S;34°43'E 23.1.97–18.5.97 600500
Yacht Club 0°8.5'S; 34° 45.5'E22.1.97–3.6.98 3 500 6 705
Usoma Beach 0°06'S;34°38'E 21.2.97–14.7.98 10 695 5 967
Karamadhan 0°07'S;34°38'E 27.2.97–15.4.98 6 250750
Otonglo Beach 0°04S;34°39.5'E 21.7.97–7.6.99 3 100 3 071
Dunga Beach 0°09'S;34°46.5'E 23.5.98–6.8.98 3 680 2 042
Kaloka Beach 0°9.5'S;34°32.5'E 17.5.98–28.5.98 2 075
Sango-Rota Beach 0°16.5'S, 34°47.5'E 7.6.97–25.3.98 2 153
30.7.99–30.9.9910 000
Kusa Beach 0°18.5'S,34°51'E 17.11.98 1 066
Nduru Beach 0°15.5'S;34°51.5'E 21.5.98979
Kendu Bay Pier 0°20'S;34°39'E 7.6.97–10.8.98 2 150
K’Owuor Pier 0°21'S;34°28'E 30.1.98 740
Homa Bay Pier 0°31'S;34°28'E 21.11.97 509
Ombogo Beach 0°28.5'S;34°30'E 29.1.98 430
Tagache Beach 0°58'S,34°6.5'E 28.1.98100
Sori-Karungu Beach 0°50'S,34°10'E 29.1.98 300
Luanda Nyamasare Beach 0°27'S,34°17'E 21.4.98508
Aram Beach 0°18'S,34°16'E 12.11.97–11.3.98 1 250
Usenge 0°03'S,34°05'E 12.11.97–16.5.98 1 250
Usigu (Uharia) Beach 0°04'S,34°9.5'E 3.2.98–16.5.98 1 750
Obenge Beach 0°13'S,34°12.5'E 16.5.98 540
Luanda Kotieno 0°18'S,34°16'E 11.3.98 250
Sio Port 0°14'N,34°02'E 12.9.97–13.1.99 1 200
Bukoma Beach 0°12'N,33°58.5'E 29.9.97–2.2.98 1 046
Nyamware Beach 0°16'S, 34°42'E13.8.98950
000
12.7.99–30.9.99 15 Ogenya Beach 0°15.5'S,34°52'E20.5.99–22.5.99471
Total41 97573 225
93
Table 3.Post-release sampling data to evaluate the impact of Neochetina weevils on water hyacinth at four sites in Lake Victoria, Kenya
Site Sampling date Fresh weight
(g) ± SE Leaf length
(cm) ± SE
Laminar area a
(cm2) ± SE
a Second youngest petiole sampled Feeding scars
± SE
Weevils/plant
± SE
Sio Port (40 m2)19.11.971685±958 137.2 ± 14.9195.4±9.7 2.5 ± 1.90.4 ± 1.4 10.3.983550±1755 77.8 ± 23.0110.2±12.5100.3 ± 9.6 1.8 ± 2.6
Bukoma Beach (15 m2)20.11.972270±935 162.9 ± 16.5178.6±0.7 2.5 ± 2.10.2 ± 0.6 10.3.98925±52875.5 ± 19.9126.8±13.0107.4 ± 28.2 2.2 ± 1.9
Police Pier (1500 m2)28.11.97251±12819.9 ± 4.849.0±5.019.4 ± 6.90.4 ± 0.5 20.5.98b
b n = 30. At all other sites n = 10.482±27131.3 ± 15.874.6±12.8138.8 ± 28.3 4.5 ± 3.9
Kendu Bay Pier (400 m2)21.11.97 1950±797 78.8 ±19.3146.8±5.6 2.9 ± 2.30.1 ± 0.3 12.3.982510±127 100.3 ±33.1124.8±13.1268.3 ± 52.4 6.0 ± 3.0
Table 4.Neochetina weevil populations on water hyacinth estimated from six sites in Lake Victoria, Kenya, May–December 1999.
Site Sampling date Mean no. of
weevils/plant a
± SE
a Mean of 10 plants per site, except for Homa Bay 18.9.99, where n=20
Mean no. of weevils by species b ± SE b Nb = Neochetina bruchi. Ne = N. eichhorniae.Range
Nb Ne
Kisumu Bay
Police Pier 6.5.99 2.5 ± 2.3 1.8 ± 0.90.7 ± 0.31–4
14.12.99 6.3 ± 4.5 5.1 ± 3.3 1.2 ± 0.60–6 Karamadhan 6.5.99 1.8 ± 2.1 1.1 ± 1.70.7 ± 0.90–6
4.12.99 3.7 ± 2.8 2.9 ± 2.30.8 ± 0.90–7 Nyakach Bay
Kusa7.5.9914.0 ± 6.714.0 ± 6.70.0 ± 0.02–22
14.12.99 3.2 ± 3.9 3.2 ± 3.90.0 ± 0.00–11 Sango Rota7.5.99 5.4 ± 4.4 3.7 ± 2.9 1.7 ± 2.31–13
15.12.99 2.9 ± 1.6 1.5 ± 1.9 1.4 ± 1.60–8 Kendu Bay
Kendu Bay7.5.99 2.4 ± 2.3 2.0 ± 1.70.4 ± 0.70–6 Pier16.12.99 2.0 ± 1.7 1.5 ± 1.60.5 ± 0.90–4 Homa Bay
Homa Bay18.9.9918.1 ± 15.315.4 ± 7.2 2.7 ± 2.30–32 Pier15.12.999.2 ± 8.6 6.5 ± 5.9 2.7 ± 3.00–32
Grand mean 6.0 ± 5.3 4.4 ± 3.1 1.6 ± 0.9
Percentage73.326.7
94
发布评论