Biocontrol Trip Logs

Lake Bisina Expedition

India Trip Report

East African Diaries

Photogallery: East African Diaries

Lake Bisina Expedition: April 5-9, 2009
by Bill Overholt

Location of Lake Bisina in Uganda
Location of Lake Bisina in Uganda

Lake Bisina
Lake Bisina

April 5: We leave Jinja at 8 AM and proceed by road to Lake Bisina. I am accompanied by Dr. Bob Copeland (entomologist employed by the International Center of Insect Physiology and Ecology, based in Jinja, Uganda, and working full time on the hydrilla project), Brian Gidudu (MSc student enrolled in a UF distance education program, and conducting research on hydrilla), and Godfrey, a water quality technician at NaFIRRI (National Fisheries Resources Research Institute, our collaborating organization in Uganda). We arrived at Kumi, a town about 10 km from Lake Bisina, around 11:30 AM, checked into the Kumi Hotel, and proceeded to Agule, a small village on the lake shore. At Agule we linked up with Dr. Fred Wanda (NaFIRRI scientist) and several other NaFIRRI staff members who had arrived at the lake one day earlier.


Construction of fish exclusion at Agule, Lake Bisina, Uganda, April, 2009
Construction of fish exclusion cages at
Agule, Lake Bisina, Uganda, April 2009

There were three objectives of the expedition; 1) to set up a fish exclusion experiment, 2) to conduct a aquatic plant diversity survey and 3) to collect hydrilla samples for DNA analyses. The purpose of the fish exclusion experiment was to measure the impact of herbivorous haplochromine fish on the performance of hydrilla. An earlier study in which Brian dissected the gut contents of fish trapped in gill nets in Lake Bisina had shown that at least four species of haplochromines consumed hydrilla. In September, 08 we made a first attempt to set up an exclusion experiment but the cages were washed away in a storm about two weeks after their installation. The new design we were using consisted of a bamboo frame covered by a coarse netting (ca. 8 mm openings). The bamboo in the lower part of the cage was partially filled with cement, and we planned to anchor the cages to the bottom of the lake with large stones. The study design included 6 cages, with three closed with netting and three open to allow entry of fish. Plant performance would be compared between the two treatments on a monthly basis to test the hypothesis that hydrilla protected from herbivory by fish would outperform hydrilla exposed to herbivory. Dr. Wanda and his team were busily constructing the cages when we arrived.


Brian Gidudu (MSc student) tossing the plant sampling tool

Brian Gidudu (MSc student) tossing the
plant sampling tool.

As the construction team did not appear to require additional help, Bob, Brian, Godfrey and I boarded a large wooden canoe and began the plant distribution survey. The plant survey involved sampling submersed macrophytes at 38 locations in the lake, which were spaced at 1 km along 6 transects. At each location, water depth was measured, and water quality measurements were made (DO, pH, temperature, conductivity, turbidity (Secchi). Water samples were collected for later analysis of phosphates, nitrates and chlorophyll A at the NaFIRRI lab. Plants were then collected from the bottom by raking 3 times with a grappling hook-like device attached to a rope. Plants were identified and scored for abundance (scale of 1-5; rare, occasional, frequent, abundant and dominant). When hydrilla was found, two tips were placed in RNAlater (an RNA preservative) for later analysis by Dean Williams. On April 5 we completed sampling 18 sites, and then returned to the hotel at sundown. The hotel was a fairly new establishment, and very clean. Power was out during much of our time there, so there was no hot water, and more importantly, no cold beer – another sacrifice in the pursuit of science.


Loading the cages on a boat for transport to Alelesi
Loading the cages on a boat for transport
to Alelesi.

April 6-8: We drove to the lake early the next morning and continued sampling for the remainder of the day, returning to cold showers and warm beers at the hotel in the late afternoon. The next morning, we sampled the final 5 sites, and then began to move the completed cages from Agule, to Alelesi, a location across the lake. Alelesi was the only spot in the lake where we had consistently found relatively dense patches of hydrilla in previous surveys. We rented a very large wooden canoe, the length of which I estimated to be 35-40 feet. The boat was large enough to transport three cages at a time, so two trips were made to transport all six cages across the lake, with each trip taking about 40 minutes each way.


Transporting the cages from Agule to Alelesi
Transporting the cages from Agule to

On the second trip across the lake, we towed a smaller canoe which we thought would be needed to place the cages in the lake. It was about 1pm by the time we were ready to drop the first cage. Cage placement consisted of having a diver (me) search for hydrilla on the bottom (ca. 3 meter deep water), and once found, having the team in the boat drop the cage in the designated area. It was easier than I had anticipated, and all the cages were in place (but not yet anchored) by about 2:30 pm. The next step was to collect a sample of hydrilla from each cage by diving to the bottom or the cage 2-3 times and grabbing handfuls, which were bagged and labeled for later inspection in Jinja. The final step was to place large rocks (estimated at ~15kg each) at diagonals about two meters from the four corners of each cage. Ropes were tied to the rocks and they were lowered to the bottom, tying the other end of the rope to the corners of the cages. Our hope was that this would hold the cages in place for 4-6 months. All afternoon we had been watching a storm in the distance, but we thought we had time to finish before it arrived. Unfortunately, we were only able to anchor one corner of one cage before the massive thunderstorm blew in from the east. The water morphed from mirror-like calmness to a sea of whitecaps within a few minutes. The wind must have been blowing 70km/hour. I was still in the water tying the anchor rope to the cage, while Bob was in the small boat yelling at me to grab ahold of the boat. I grabbed the back of the canoe and was quickly towed the 50 meters to the shore. The people in the larger, motorized boat later told us that they tried to come get us, but the boat was unable to make any headway against the wind. We arrived on shore, pulled the canoe up as far as we could, and headed for a large tree under which we shivered for the next 30-40 minutes, until the worst of the storm passed. I had no raincoat, so Bob cut a hole in a large plastic bag we had for collecting hydrilla samples, and pulled it over my torso.


Brian Gidudu (left under umbrella) and Bill Overholt (in bag) waiting out a storm
Brian Gidudu (left under umbrella) and Bill
Overholt (in bag) waiting out a storm.

It was still raining and windy, but visibility had returned. Unfortunately, we could only spot 5 of the six cages. We directed the small canoe to tie one rock to each of the 5 cages, and then we headed back to Agule, and on to Kumi, arriving exhausted just before dark. The only bright side was that the electricity had been on most of the day, so there was luke warm water for showers, and the beer was at nearly an acceptable temperature.


An exclusion cage in place
An exclusion cage in place

Floating islands (suds) in Lake Bisina after a storm
Floating islands (suds) in Lake Bisina after
a storm.

The next morning we headed back to Alelesi to assess the damage. Five cages were found, moved over hydrilla and anchored, and new samples taken. The sixth cage (luckily an open one) was nowhere to be found. A floating island had blown into the area and may have pushed over the cage and covered it. Floating islands, locally called ‘suds’, break-off from the mainland during the rainy season and move around the lake with the winds; their size can range from a few square meters to more than one acre. Since the missing cage was open, we simple selected a sixth location, and marked it using a GPS. We completed the work around noon and drove back to Jinja. Brian will return to Bisina in two weeks to see whether the cages have stood up.

April 9: I went to the NaFIRRI lab in the morning and helped Brian take measurements on 10 randomly selected hydrilla stems from each of the six cage collections. Variables included: 1) length of main stem, 2) status of the tip of the mainstem (broken or intact), 3) total length, 4) number of internodes on the main stem, 5) status of whorls on the mainstem (eaten or intact), 6) total number of tips, 7) number of intact tips. In addition a schematic diagram of each stem was drawn. Returned to hotel, ate lunch and then drove with Bob to Entebbe where we would catch a flight to Bujumbura, Burundi the next day.