Kootenai River Nutrient Dosing System and N-P Consumption: Year 2008.

Abstract

In early 2006 we designed and built low energy consumption, pump-operated system, for dosing of the liquid nutrient in the summer 2006 season. This operated successfully, and the system was used again during the 2007 and 2008 seasons for dosing. During the early winter period, 2008, laboratory tests were made of the liquid nutrient pump system, and it was noted that small amounts of air were being entrained on the suction side of the pump, during conditions when the inlet pressure was low. It was believed that this was the cause of diurnal fluctuations in the flow supplied, characteristic of the 2007 year flow data. Replacement of '0' rings on the inlet side of the pumps was the solution to this problem, and when tested in the field during the summer season, the flow supplied was found to be stable. A decision was made by the IKERT committee at the meeting of 20th to 21st May 2008 (held in Coeur d'Alene, Idaho) to use an injection flow rate of liquid fertilizer (polyammonium phosphate 10-34-0) to achieve a target phosphorus concentration of 3.0 {micro}g/L, after complete mixing in the river. This target concentration was the same as that used in 2006 and 2007. The proposed starting date was as early as possible in June 2008. Plans were made to measure the dosing flow in three ways. Two of the three methods of flow measurement (1 and 2 below) are inter-dependent. These were: (1) Direct measurement of flow rate by diverting dosing flow into a 1000 mL volume standard flask. The flow rate was computed by dividing the flask volume by the time required to fill the flask. This was done a few times only during the summer period. (2) Adjusting the flow rate reading of the Gamma dosing pump using the 'calibration' function to achieve agreement with the flow rate computed by method 1 above. (3) Direct measurement by electrical signal from conductive fluid passing through a magnetic field (Seametrics meter, as used in previous years). Values were recorded every 4 minutes by a data-logger. This instrument has been shown to be reliable, and in agreement with method 1 to within the expected uncertainty (within 2%). Liquid nutrients were delivered to the site in late May, and system testing was done the same day. High concentrations of suspended sediment in the water column, as indicated by shallow Secchi depth readings, were present in the river in the last part of May, into June. A plan was made to delay the start up of nutrient addition to a date later than 1st June, because the aquatic productivity was almost certainly to be compromised by insufficient light availability. Daily monitoring of Secchi depths was done, showing declining turbidity in early June. A decision was made to start the system on 15th June, by which time conditions were good.