Nutrient and productivity field trips report
High water 5th – 16th May 2015, low water 16th – 26th August 2015
May trip: Geraldine Taylor1,2, Richard Peel1,2,3 (PhD students), Prof. Olaf Weyl1 (Principal Scientist), Dr Jackie Hill4 (Research Officer), Dr Michelle Jackson5 (Post Doctoral Fellow)
August trip: Geraldine Taylor1,2, Richard Peel1,2,3 (PhD students), Ann Wu1 (Intern)
1South African Institute for Aquatic Biodiversity (SAIAB), Grahamstown, South Africa
2Department of Ichthyology and Fisheries Science, Rhodes University, South Africa
3NNF/EU Community Conservation Fisheries in KAZA Project, Katima Mulilo, Namibia
4Department of Zoology & Entomology, Rhodes University, South Africa
5Centre for Invasion Biology, Department of Zoology and Entomology, University of Pretoria
To measure the nutrients and productivity of the Zambezi, Kavango and Kwando rivers and Lake Liambezi during high water and low water periods.
Aim: To characterise the nutrient concentrations of the Zambezi, Kavango and Kwando Rivers and Lake Liambezi.
Objectives: Compare the nutrient concentrations at two sites in each system, at high water and low water periods.
Characterise the nutrient concentrations spatially, encompassing both main channel and backwater sites in the rivers, and the main lake and river inlets in Lake Liambezi.
Using an Ion Selective Electrode, ammonium and nitrate ion concentrations were measured in replicates of five, at a number of sights in each system. Water quality parameters such as temperature, conductivity, total dissolved solids, pH and turbidity, GPS coordinates, and water depth were measured at all sites.
Geraldine and Richard calibrating the Ion Specific Electrode before measuring ammonium and nitrate concentrations at Lake Liambezi.
On the Zambezi River in the main channel and a backwater, nitrates were higher in the flood compared to low water, while ammonium was higher in low water.
On the Kavango River, nitrates did not change with season, while in the Kwetche side channel ammonium was higher in the low water.
In the Kwando River and on Lake Liambezi both nitrates and ammonium were higher in August compared to in May.
In the Zambezi River, main river and side channel sites had higher nutrient concentrations (0.32 mg/l NO3, 0.13 mg/l NH4) than backwater sites (0.1 mg/l NO3, 0.11 mg/l NH4).
In the Kavango River, nutrient concentrations were very low (0.1 mg/l NO3 and NH4) in the upstream section near KIFI in the main channel and in the backwaters. Nutrient levels increased with animal inputs from backwaters draining the floodplain, resulting in higher ammonium levels in both backwater and main channel sites around Kwetche and downstream towards the Botswana border. The most nutrient rich site of the Kwetche backwater measured 0.42 mg/l NH4. Nitrate levels did not change between the KIFI and the Kwetche area.
In the Kwando River upstream of Malyo, nutrient concentrations in the main channel and backwaters were low (0.32 mg/l NO3 and NH4). Concentrations increased downstream of Malyo, and were the highest in Nakatwa channels (0.53 mg/l NO3, 0.3 mg/l NH4). Within Nakatwa, backwaters had lower nutrient concentrations (0.42 mg/l NO3, 0.35 mg/l NH4) than the slowly flowing channels.
On Lake Liambezi higher nitrate concentrations (2.05 – 3.33 mg/l) were measured in the sites closer to land (eg Chobe River mouth, Bukalo Channel and Muyako launch site), compared to those in the main lake (including Linyanti 1.4 – 1.9 mg/l). In addition nitrate concentrations were positively correlated with turbidity. Ammonium concentrations ranged from 0.46 to 0.74 mg/l throughout the lake, with no clear trends observed.
Aim: Measure the primary productivity of phytoplankton in all systems during high water and low water periods.
Filtered water from the site was funnelled into ten clear and ten blackened 2.5 l glass bottles. This constituted five replicates, two clear and two dark bottles per replicate. All bottles were spiked with 2380 µl/l enriched carbon (NaHCO3), and enriched nitrogen (5 light and 5 dark bottles with 435 µl/l KNO3, and 5 light and 5 dark bottles with 230 µl/l NH4Cl) using a 1000 µl pipette. Each replicate (consisting of two pairs of light and dark bottles) was placed within 30 cm of the water surface, anchored on a buoy, and left to incubate for six hours. All five replicated were within 100 m of each other. At the location of each replicate GPS coordinates and water quality parameters were taken, and a HOBO temperature logger was secured to one bottle, which measured water temperature every five minutes. Two HOBO temperature and light loggers were placed at the same water depth, in the vicinity of all of the replicates, to record these variables every five minutes.
Five natural particulate organic matter (POM) samples were taken at each site (one per replicate).
After six hours of incubation, a measured amount of water from each bottle was filtered through glass fiber filters using a filtration manifold and hand held vacuum pump. These enriched POM samples were then dried in a drying oven at 50 °C, and sent to the Stable Isotope Laboratory, Mammal Research Institute (MRI), University of Pretoria for analysis.
In total 400 samples have been sent to the laboratory for analysis (50 per system per field trip). Results will quantify the ratio of heavy to light carbon and nitrogen isotopes for the natural and enriched samples. From these values, the productivity of the phytoplankton present in each system can be calculated, in relation to the amount of enriched carbon and nitrogen added to each bottle, and the amount of each isotope taken up (measured by filtering) in the dark and light bottles.
Left: Incubating dark and light bottles (half of one replicate) spiked with enriched carbon and nitrogen (nitrate or ammonium), right: Richard filtering water from a bottle spiked with enriched carbon and nitrogen at Nakatwa on the Kwando River.
This information on nutrient contents and primary productivity is useful for characterising and comparing these systems. An understanding of these properties will also contribute towards explaining differences in fish diets, feeding and growth, and food web structure. Furthermore baseline data on these systems is lacking, and will be useful for future comparisons.
Thanks to the EU-NNF Fisheries Project for funding this research. Thanks to Malyo Wilderness Camp for allowing us to stay at your beautiful camp for free in May. Thanks to all of the scientists who participated. Thanks Denis Tweddle for letting us stay with you.