Arrow Lakes Reservoir Nutrient Restoration Program

Author: M. Bassett, E.U. Schindler, R. Fox, L. Vidmanic, T. Weir, D. Sebastian, K. Peck

To address the ultra-oligotrophic status of these systems, a bottom-up approach was taken with the addition of nutrients (nitrogen and phosphorus in the form of liquid fertilizer) to increase the production of Daphnia spp., a main food source for kokanee, which in turn is a keystone species in these lakes. Nutrient addition to the Arrow Lakes Reservoir began in 1999. The objectives of the nutrient restoration program are outlined in the Fish and Wildlife Compensation Programs Large Lakes Plan. The program follows habitat based actions, one of which is to apply seasonally adjusted nutrients to mimic natural inputs of nutrients to sustain productivity at levels sufficient to support fisheries and ecosystem objectives.

The average April to October mean daily outflow in Arrow Lakes Reservoir in 2016 was 1171 m3/s, which was slightly above the average long term (19972016) daily mean. Overall, flows were 89% of normal from April to September.
Arrow Lakes Reservoir began to stratify in June, and likely at some stations as early as May, then displayed warming surface temperatures through July and August. As in previous years, summer stratification occurred with the epilimnion becoming more clearly defined in late summer and early fall. Stratification was maintained until as late as November at some stations. In 2016, hypolimnetic temperatures ranged from 3.54°C throughout the year, which is comparable to previous years.
Seasonally, conductivity was highest in the spring for both Upper and Lower Arrow, coinciding with freshet.
Secchi disc measurements in Arrow in 2016 show a typical seasonal pattern of decreasing transparency associated with the spring phytoplankton bloom and freshet, followed by an increase in transparency as the bloom and freshet gradually abates by the late summer.
In 2016 in the integrated 020 m samples, 36.9% of total phosphorus (TP) values and 64.4% of total dissolved phosphorus (TDP) values were reported at less than the RDL. The annual mean for monthly TP values in 2016 for Upper Arrow was 3.2 μg/L, slightly lower than the 19972016 mean of 3.3 μg/L. The mean TP for Lower Arrow was 2.6 μg/L, lower than the Lower Arrow basin 19972016 mean of 3.3 μg/L. The annual mean for monthly TN values in 2016 for Upper Arrow was 161.6 μg/L, lower than the 19972016 mean of 189.8 μg/L. Lower Arrow was 136.7 μg/L, lower than the Lower Arrow basin 19972016 mean of 183.3 μg/L. The annual mean for monthly N:P ratios in 2016 for Upper Arrow was 58, which was slightly higher than the 19972016 mean of 55. Lower Arrow was 38, or slightly lower than the Lower Arrow basin 19972016 mean of 41. In 2016, N:P ratios in Upper Arrow ranged from 26 (AR2 in late June) to 88 (which was observed at AR1 in May). In Lower Arrow, N:P ratios ranged from 14 (AR8 in June) to 58 (AR8 in April).
Total monthly phytoplankton mean abundance by basin ranged from 1091 cells/mL in April in Upper Arrow to 6508 cells/mL in the Narrows in October. Typical of April and May samples, the taxonomic group that dominated the total abundance was chryso- & cryptophytes. Similar to previous summer and early fall samples, the dominant taxonomic group was the Bacillariophytes. Biovolume followed the same patterns as abundance: biovolume was lowest in Upper Arrow in April at 0.100 mm3/L, and highest in the Narrows in October.
In 2016, the biovolume of phytoplankton fluctuated over the course of the sampling period. The phytoplankton community followed a typical trend of building over the course of the growing season, and tapering off as fall conditions of lower nutrients, colder temperatures and shorter photoperiods set in. In 2016 for all basins, the edible mean increased from 2015, and was lower than the long term mean. The annual biovolume of inedible phytoplankton was also slightly below the long-term means for all basins. The lower 2016 results for the edible community may be attributed to the higher zooplankton biomass and that zooplankton were first observed slightly earlier in the year than on average (see zooplankton results).
Twenty species of macrozooplankton were identified in the samples over the course of the study, with copepods such as Leptodiaptomus ashlandi, Epishura nevadensis and Diacyclops bicuspidatus thomasi, and the cladocerans Daphnia galeata mendotae and Bosmina longirostris being the most numerous. Total average zooplankton density in the Upper Arrow in 2016 was dominated by copepods, which comprised 84% of total zooplankton density with a count of 10.87 individuals/L. Other cladocerans comprised 9% of zooplankton density with 1.14 individuals/L, while Daphnia spp. contributed only 8% with 0.98 individuals/L. In the Lower Arrow, copepods contributed to 71% of total zooplankton density (15.63 individuals/L), Daphnia spp. 16 % (3.57 individuals/L) and cladocerans other than Daphnia spp. 13% (2.91 individuals/L).
Copepods were the main contributor to the overall zooplankton population throughout the sampling season, with Daphnia present in each month from April to November, peaking in August or September and maintaining a population through November. This pattern occurred in both Upper and Lower Arrow in 2016 (Figure 22 for density and Figure 23 for biomass). In 2016 copepods dominated the abundance and biomass at the beginning of the sampling season, from April through July, while Daphnia dominated in biomass from July through October.
Total zooplankton density was higher in Lower Arrow than Upper Arrow in 2016: a pattern that is repeated in each studied year. Total zooplankton biomass was almost three times higher in Lower Arrow than in Upper Arrow. Like density, biomass has always been higher in Lower Arrow then Upper Arrow since 1997.
The first appearance by month and by basin of Daphnia for each year is shown in Figure 24. On average, Daphnia begin to appear in May in Upper Arrow samples, and June in Lower Arrow samples. 2016 was an early year: in Upper Arrow, Daphnia were first observed in the April samples and in Lower Arrow Daphnia first appeared in May samples. Typically, Upper Arrow Daphnia are observed one to two months sooner or simultaneously with Lower Arrow, aside from in 2005 and 2011.
The 2016 monthly results are shown in Figure 25 for Upper and Lower Arrow density and biomass. Density results varied from station to station in both Lower and Upper Arrow. Biomass results were similar among stations during all months in Upper Arrow, while in Lower Arrow biomass results were similar among stations from April to June, but varied between stations from July through October. The highest density was recorded in July and the highest biomass in August, which were both in Lower Arrow at station AR6.
Density of Mysis diluviana (mysids) has fluctuated over the course of the studied years. Compared to the previous year, densities of mysids in 2016 increased in Upper Arrow but decreased in Lower Arrow. The peak density in 2016 in Upper Arrow occurred in June at station AR2 with 861.50 ind/L, mainly due to an increased number of juveniles. In Lower Arrow the peak density occurred in July at station AR8 with 352 ind/L, also due mainly to an increased number of juveniles. Compared to 2015, average biomass in Upper Arrow increased while in Lower Arrow biomass decreased.
Pool elevation during the kokanee October 14, 2016 survey period averaged 427.75 m (12.5 m below full pool). Pool elevation at the time of the 2016 survey was 6.1 m below the average for fall surveys, even lower than the 2015 survey which was the lowest since 2001. The total area of pelagic habitat (defined as >20 m depth) was estimated at 193 km2 in Upper Arrow and 91 km2 in Lower Arrow, or approximately 2% and 5.5% lower than the respective long-term averages for fall surveys, respectively.
A total of 579 kokanee were captured at the six standard trawl stations in 2016: 268 from Upper Arrow and 311 from Lower Arrow. The Upper Arrow trawl catch was 125, 112, and 31 for ages 0, 1, and 2, respectively. In Lower Arrow, the trawl sampling produced catches of 135, 101, and 75 for ages 0, 1, and 2, respectively. Age-specific length frequencies for spawning kokanee from Hill Creek in Upper Arrow and from Deer, Taite, and Mosquito Creeks in Lower Arrow were consistent with trawl age-specific length frequencies for Upper and Lower Arrow, respectively. Length frequency distributions show a single mode of spawners returned to Hill Creek in for the previous five consecutive years, including 2016. The pattern of a single age class dominating age at maturity appears to have changed in 2016, as otolith ageing revealed more age 2 than age 3 spawners at 58% and 42% respectively for Hill Creek spawners.
A return of 74,751 spawners to Hill Creek in 2016 represented 48% of average for the nutrient addition period and was down from 56% of average in 2015. A return of only 8,141 spawners to all other index tributaries in Upper Arrow in 2016 represented only 30% of the fertilization era average, but double the 2015 return to these same tributaries.
Reservoir fry densities in 2016 ranged from 171871 fish.ha-1, and averaged 298 fish.ha-1 in Upper Arrow and 464 fish.ha-1 in Lower Arrow. Annual hydroacoustic estimates for kokanee during fall surveys have ranged from 520 million and averaged 10.3 million since nutrient additions began in 1999. In 2016, the total kokanee abundance remained above average and was estimated at 11.8 million (10.213.4 M).
In 2016, kokanee biomass density estimates increased substantially to 6.3 kg/ha in Upper Arrow and 7.1 kg/ha in Lower Arrow, slightly above the nutrient era average for both basins and the highest estimate since 2009 in Upper Arrow and 2010 in Lower Arrow.
In 2016, the egg deposition estimate for the spawning channel was 4.5 million based on 41,344 adults or 19,432 females (47.0% female) with a net fecundity of 283 eggs/female. The 2016 egg deposition was lower than the previous three years and a decrease in egg to fry survival in 2016 resulted in the below target fry emigration of 2.6 million from the HCSC.
survival trend has been variable with good kokanee survival in 2001, 2006, 2008 and 2009 and poor survival in 2004, 2007, 2012, and 2015. The annual survival index was the highest on record in 2006 at nearly 2 SD above average, driven primarily by excellent survival from age 0 to 1 but also slightly above average survival from age 1 to 2. The following year, survival plunged dramatically for both age 01 and age 12 to approximately 1 SD below the mean. The annual survival was the worst on record in 2012, when both the age 01 and age 12 values were >1 SD below the mean. Kokanee survival has remained below average since 2012 although returned to near average in 2016.

The average April to October mean daily outflow in Arrow Lakes Reservoir in 2016 was 1171 m3/s, which was slightly above the average long term (19972016) daily mean. Overall, flows were 89% of normal from April to September.
Arrow Lakes Reservoir began to stratify in June, and likely at some stations as early as May, then displayed warming surface temperatures through July and August. As in previous years, summer stratification occurred with the epilimnion becoming more clearly defined in late summer and early fall. Stratification was maintained until as late as November at some stations. In 2016, hypolimnetic temperatures ranged from 3.54°C throughout the year, which is comparable to previous years.
Seasonally, conductivity was highest in the spring for both Upper and Lower Arrow, coinciding with freshet.
Secchi disc measurements in Arrow in 2016 show a typical seasonal pattern of decreasing transparency associated with the spring phytoplankton bloom and freshet, followed by an increase in transparency as the bloom and freshet gradually abates by the late summer.
In 2016 in the integrated 020 m samples, 36.9% of total phosphorus (TP) values and 64.4% of total dissolved phosphorus (TDP) values were reported at less than the RDL. The annual mean for monthly TP values in 2016 for Upper Arrow was 3.2 μg/L, slightly lower than the 19972016 mean of 3.3 μg/L. The mean TP for Lower Arrow was 2.6 μg/L, lower than the Lower Arrow basin 19972016 mean of 3.3 μg/L. The annual mean for monthly TN values in 2016 for Upper Arrow was 161.6 μg/L, lower than the 19972016 mean of 189.8 μg/L. Lower Arrow was 136.7 μg/L, lower than the Lower Arrow basin 19972016 mean of 183.3 μg/L. The annual mean for monthly N:P ratios in 2016 for Upper Arrow was 58, which was slightly higher than the 19972016 mean of 55. Lower Arrow was 38, or slightly lower than the Lower Arrow basin 19972016 mean of 41. In 2016, N:P ratios in Upper Arrow ranged from 26 (AR2 in late June) to 88 (which was observed at AR1 in May). In Lower Arrow, N:P ratios ranged from 14 (AR8 in June) to 58 (AR8 in April).
Total monthly phytoplankton mean abundance by basin ranged from 1091 cells/mL in April in Upper Arrow to 6508 cells/mL in the Narrows in October. Typical of April and May samples, the taxonomic group that dominated the total abundance was chryso- & cryptophytes. Similar to previous summer and early fall samples, the dominant taxonomic group was the Bacillariophytes. Biovolume followed the same patterns as abundance: biovolume was lowest in Upper Arrow in April at 0.100 mm3/L, and highest in the Narrows in October.
In 2016, the biovolume of phytoplankton fluctuated over the course of the sampling period. The phytoplankton community followed a typical trend of building over the course of the growing season, and tapering off as fall conditions of lower nutrients, colder temperatures and shorter photoperiods set in. In 2016 for all basins, the edible mean increased from 2015, and was lower than the long term mean. The annual biovolume of inedible phytoplankton was also slightly below the long-term means for all basins. The lower 2016 results for the edible community may be attributed to the higher zooplankton biomass and that zooplankton were first observed slightly earlier in the year than on average (see zooplankton results).
Twenty species of macrozooplankton were identified in the samples over the course of the study, with copepods such as Leptodiaptomus ashlandi, Epishura nevadensis and Diacyclops bicuspidatus thomasi, and the cladocerans Daphnia galeata mendotae and Bosmina longirostris being the most numerous. Total average zooplankton density in the Upper Arrow in 2016 was dominated by copepods, which comprised 84% of total zooplankton density with a count of 10.87 individuals/L. Other cladocerans comprised 9% of zooplankton density with 1.14 individuals/L, while Daphnia spp. contributed only 8% with 0.98 individuals/L. In the Lower Arrow, copepods contributed to 71% of total zooplankton density (15.63 individuals/L), Daphnia spp. 16 % (3.57 individuals/L) and cladocerans other than Daphnia spp. 13% (2.91 individuals/L).
Copepods were the main contributor to the overall zooplankton population throughout the sampling season, with Daphnia present in each month from April to November, peaking in August or September and maintaining a population through November. This pattern occurred in both Upper and Lower Arrow in 2016 (Figure 22 for density and Figure 23 for biomass). In 2016 copepods dominated the abundance and biomass at the beginning of the sampling season, from April through July, while Daphnia dominated in biomass from July through October.
Total zooplankton density was higher in Lower Arrow than Upper Arrow in 2016: a pattern that is repeated in each studied year. Total zooplankton biomass was almost three times higher in Lower Arrow than in Upper Arrow. Like density, biomass has always been higher in Lower Arrow then Upper Arrow since 1997.
The first appearance by month and by basin of Daphnia for each year is shown in Figure 24. On average, Daphnia begin to appear in May in Upper Arrow samples, and June in Lower Arrow samples. 2016 was an early year: in Upper Arrow, Daphnia were first observed in the April samples and in Lower Arrow Daphnia first appeared in May samples. Typically, Upper Arrow Daphnia are observed one to two months sooner or simultaneously with Lower Arrow, aside from in 2005 and 2011.
The 2016 monthly results are shown in Figure 25 for Upper and Lower Arrow density and biomass. Density results varied from station to station in both Lower and Upper Arrow. Biomass results were similar among stations during all months in Upper Arrow, while in Lower Arrow biomass results were similar among stations from April to June, but varied between stations from July through October. The highest density was recorded in July and the highest biomass in August, which were both in Lower Arrow at station AR6.
Density of Mysis diluviana (mysids) has fluctuated over the course of the studied years. Compared to the previous year, densities of mysids in 2016 increased in Upper Arrow but decreased in Lower Arrow. The peak density in 2016 in Upper Arrow occurred in June at station AR2 with 861.50 ind/L, mainly due to an increased number of juveniles. In Lower Arrow the peak density occurred in July at station AR8 with 352 ind/L, also due mainly to an increased number of juveniles. Compared to 2015, average biomass in Upper Arrow increased while in Lower Arrow biomass decreased.
Pool elevation during the kokanee October 14, 2016 survey period averaged 427.75 m (12.5 m below full pool). Pool elevation at the time of the 2016 survey was 6.1 m below the average for fall surveys, even lower than the 2015 survey which was the lowest since 2001. The total area of pelagic habitat (defined as >20 m depth) was estimated at 193 km2 in Upper Arrow and 91 km2 in Lower Arrow, or approximately 2% and 5.5% lower than the respective long-term averages for fall surveys, respectively.
A total of 579 kokanee were captured at the six standard trawl stations in 2016: 268 from Upper Arrow and 311 from Lower Arrow. The Upper Arrow trawl catch was 125, 112, and 31 for ages 0, 1, and 2, respectively. In Lower Arrow, the trawl sampling produced catches of 135, 101, and 75 for ages 0, 1, and 2, respectively. Age-specific length frequencies for spawning kokanee from Hill Creek in Upper Arrow and from Deer, Taite, and Mosquito Creeks in Lower Arrow were consistent with trawl age-specific length frequencies for Upper and Lower Arrow, respectively. Length frequency distributions show a single mode of spawners returned to Hill Creek in for the previous five consecutive years, including 2016. The pattern of a single age class dominating age at maturity appears to have changed in 2016, as otolith ageing revealed more age 2 than age 3 spawners at 58% and 42% respectively for Hill Creek spawners.
A return of 74,751 spawners to Hill Creek in 2016 represented 48% of average for the nutrient addition period and was down from 56% of average in 2015. A return of only 8,141 spawners to all other index tributaries in Upper Arrow in 2016 represented only 30% of the fertilization era average, but double the 2015 return to these same tributaries.
Reservoir fry densities in 2016 ranged from 171871 fish.ha-1, and averaged 298 fish.ha-1 in Upper Arrow and 464 fish.ha-1 in Lower Arrow. Annual hydroacoustic estimates for kokanee during fall surveys have ranged from 520 million and averaged 10.3 million since nutrient additions began in 1999. In 2016, the total kokanee abundance remained above average and was estimated at 11.8 million (10.213.4 M).
In 2016, kokanee biomass density estimates increased substantially to 6.3 kg/ha in Upper Arrow and 7.1 kg/ha in Lower Arrow, slightly above the nutrient era average for both basins and the highest estimate since 2009 in Upper Arrow and 2010 in Lower Arrow.
In 2016, the egg deposition estimate for the spawning channel was 4.5 million based on 41,344 adults or 19,432 females (47.0% female) with a net fecundity of 283 eggs/female. The 2016 egg deposition was lower than the previous three years and a decrease in egg to fry survival in 2016 resulted in the below target fry emigration of 2.6 million from the HCSC.
survival trend has been variable with good kokanee survival in 2001, 2006, 2008 and 2009 and poor survival in 2004, 2007, 2012, and 2015. The annual survival index was the highest on record in 2006 at nearly 2 SD above average, driven primarily by excellent survival from age 0 to 1 but also slightly above average survival from age 1 to 2. The following year, survival plunged dramatically for both age 01 and age 12 to approximately 1 SD below the mean. The annual survival was the worst on record in 2012, when both the age 01 and age 12 values were >1 SD below the mean. Kokanee survival has remained below average since 2012 although returned to near average in 2016.