Summary

Hydrologic and water-quality conditions were monitored in the newly restored Wood River Wetland in the upper Klamath River basin, Oregon, from May 2003 to September 2005 to establish baseline conditions as a prelude to development of future adaptive management options. Nitrogen, phosphorus, and instantaneous measures of field parameters (water temperature, dissolved oxygen, pH, and specific conductance) were monitored approximately monthly during the growing season at surface- and ground-water sites in the North and South Units of the Wood River Wetland. Limited dissolved organic carbon and selected dissolved metals data also were collected.

Solute concentrations in the wetland were higher than in adjacent streams or Agency Lake. Potential sources of nutrients in the wetland include releases from peat soils and decomposition of organic matter, including emergent, submergent, and floating macrophytes, algae, and waterfowl feces. Another factor contributing to high solute concentrations is summer evaporation and evapotranspiration, which averaged about 3.7 feet of water per year. The highest surface-water concentrations for most nutrients occurred in late summer, when water volumes were lowest. Evapoconcentration helped raise nutrient, carbon, and dissolved ion concentrations to exceedingly high values. This effect was most pronounced in the deeper South Unit, where specific conductance (SC) values reached 2,500 microsiemens per centimeter (µS/cm) in September 2005. During the three year study, the median concentrations of total nitrogen (TN) and total phosphorus (TP) in the South Unit reached about 18,000–36,000 and 18,000–26,000 micrograms per liter (µg/L), respectively. Most of the TN and TP was in dissolved form, primarily as dissolved organic nitrogen or ammonium, and soluble reactive phosphorus (SRP). Nutrient concentrations in surface water generally were much lower in the North Unit than in the South Unit, which has deeper peat soils, although one set of samples collected in December 2003 from the North Unit contained elevated concentrations of nitrate.

Spring and early summer increases in dissolved ammonium and SRP outpaced chloride, a conservative tracer indicative of evapoconcentration. This indicates that biogeochemical processes other than evapoconcentration are responsible for the elevated nutrient concentrations in the wetland. Lower SRP and TP levels occurred in autumn following the annual flood irrigation of the North Unit and a return to cooler and wetter winter conditions. The SRP and TP concentrations decreased at a rate that was faster than dilution, indicating that inorganic sorption may have occurred in the wetland, possibly involving iron, manganese, or calcium.

Because of the elevated ammonium concentrations (and relatively high water temperature and moderate pH), some highly toxic un-ionized ammonia (NH₃) developed in the wetland surface waters. The median ammonium concentrations in the North and South Units were 380 and 510 µg-N/L, respectively, which equates to NH₃ concentrations of less than about 30 µg-N/L at the ambient temperature and pH. The maximum total ammonia concentration of 2,020 µg/L in the South Unit during August 2004 equates to 390 µg-N/L un-ionized ammonia (NH₃) at the ambient water temperature and pH (25°C and 8.1 units, respectively). This NH₃ level might stress fish, including endangered Lost River and shortnose sucker fish, which have respective 96-hour LC₅₀ (median lethal concentration for one-half the test population) values of 780 and 530 µg/L NH₃ (Saiki and others, 1999). Although a fish screen is used at the wetland inlet structure, water with elevated NH₃ discharged to Wood River or Sevenmile Canal during pumping might affect fish in the immediate vicinity of the pump outlet station, depending on a number of factors, including the total ammonia concentration, pH, water temperature, and streamflow (amount of dilution) in the receiving stream (Wood River or Sevenmile Canal).

Seasonal increases in dissolved nutrients and minerals result from mineralization of peat soils. Previous studies of soil cores indicated land subsidence of about 4–5 feet, resulting in carbon, nitrogen, and phosphorus losses of about 36, 31, and 18 percent of their initial (predrainage) mass, respectively. Peat-associated nitrogen likely contributed to the high ammonium concentrations (22,900–36,500 µg‑N/L) in the 26–28 ft well in the North Unit. Elevated nitrogen and phosphorus concentrations (about 6,000–7,000 µg/L of ammonium and SRP) were measured in water samples from the five continuously flowing artesian wells. Biological activity was apparent in the wetland, including development of algae and floating and submerged aquatic plants, which boosted median total Kjeldahl nitrogen levels to 10,600 µg-N/L and TP to about 7,800 µg-P/L in the South Unit. The rapid decrease in ammonium, nitrate, and SRP during the latter part of the growing season (August and September) indicated biological uptake from algae, macrophytes, and other wetland vegetation. Photosynthesis by aquatic plants sometimes produced supersaturated midday dissolved oxygen concentrations in surface waters (as much as 310 percent saturation) and pH up to 9.2 units during summer.

Dissolved oxygen concentrations at wetland surface-water sites decreased each summer, and minimum values were less than 0.5 mg/L each year. Decomposition of peat soils and vegetation likely contributed to the decreased dissolved oxygen levels measured in the wetland surface and ground waters. Some areas of the wetland supported sulfate reduction indicative of anaerobic conditions. Anaerobic soil conditions in wetland sediments act to halt the oxidation and decomposition of peat soils, and are a natural condition in functioning and healthy wetlands.

In-situ mesocosm dome chamber experiments conducted in the South Unit during June and August 2005 indicated active demand for dissolved oxygen (0.15–1.0 milligrams of oxygen per liter per hour [mg-O₂/L/h]). Positive flux of ammonium and SRP from the bed sediment also was measured during the chamber experiments. This release may be attributed to the upward movement of ground water through nutrient-rich decomposing peat soils (for nitrogen and phosphorus), and desorption from iron or manganese in the soil under anaerobic conditions (for phosphorus). The chamber experiments also confirmed active denitrification, but rates were low due to low nitrate levels in the incubation water (wetland surface water). The low nitrate probably results from a combination of denitrification in the wetland, uptake by wetland vegetation, and perhaps from a microbial process termed dissimilatory nitrate reduction to ammonium (DNRA). The concentration of dissolved ammonium increased in the chambers with additions of nitrate and carbon, which indicates the possible occurrence of DNRA. Overall, these experiments showed active nutrient cycling by microbial populations in the wetland sediments.

Surface-water levels and standing surface-water volumes in the Wood River Wetland reached a maximum in early spring, inundating 80–90 percent of the wetland. Minimum water levels occurred in August through November, when the South Unit was 10 percent inundated and the North Unit was nearly dry. Shallow ground-water levels followed a trend similar to surface-water levels and indicated a strong upward gradient. The hydrologic conditions and water levels measured during this study were influenced by the relatively dry conditions in 2003–05, and may not be representative of conditions during wet years. Additionally, water levels may not get as low considering that current management includes irrigation in late July to maintain a greater area of saturated soils.

Monthly water budgets were developed individually for the North and South Units of the Wood River Wetland and then summed to produce a water budget over a 2 year period (2004 and 2005 water years). The budget was based on estimates of inflows from precipitation, ground water, and surface water applied for irrigation; outflows to open-water evaporation and evapotranspiration due to emergent vegetation; and changes in surface-water and ground-water storage. Precipitation was the largest component of inflow for the entire wetland, representing 43 percent of the total inflow. Precipitation values were obtained from a nearby automated agricultural weather station. Inflow from regional ground-water discharge accounted for 40 percent of the total and was estimated using Darcy’s Law. Reported inflow from applied water for irrigation from adjacent surface-water bodies contributed 12 percent of the total inflow. Combined inflows from ground-water seepage through dikes and discharge from five artesian wells represented 5 percent of the total inflow. Outflows from the wetland consisted of open-water evaporation (estimated using an empirical evaporation equation) and evapotranspiration from emergent vegetation (estimated from energy balance studies of nearby wetlands) which accounted for 64 and 36 percent of the total outflow, respectively. Outflows exceeded inflows by about 22 percent over the 2-year period. Changes in surface-water and ground-water storage during this period were determined by the change in surface- and ground-water levels and amounted to losses of only 1 and 2 percent relative to the total inflow, respectively. A water-budget residual consisting of the errors in measurement or estimation of all water-budget components and the sum of any unestimated components indicated a water deficit of 19 percent relative to the total inflow. The monthly patterns in water-budget residuals for the wetland resembled a soil-moisture signal, which, although not measured, could be a significant part of the water budget.

The distribution of inflow between the North and South Units generally was proportional to their area. The primary exception was applied irrigation, which entered entirely into the North Unit. Open-water evaporation was a greater proportion of the outflow in the South Unit relative to the North Unit as a consequence of the South Unit’s large persistent areas of open water.

Inundation models developed for this study can be used to guide future water management of the wetland to reduce further oxidation of peat soils while providing shallow habitats for wetland plant seed germination. Given the current conditions, higher water levels in the North Unit and lower water levels in the South Unit (achieved through pumping) may be needed to achieve both of these management goals. Additional flushing of the wetland by flood-irrigation and pumping may be required to dilute the sometimes exceptionally high levels of dissolved salts and dissolved organic carbon in the wetland surface waters that may impede seed germination in the South Unit. Focused studies could determine the particular salt tolerances and light requirements for successful germination and growth of wetland plant species.

Determining the appropriate water level to minimize oxidation of soils while promoting seed germination is challenging due to the presence of relatively deep canals, patchy distribution of swales and ponds, and areas of subsided soils. Future studies of the various managed wetlands around Agency and Upper Klamath Lakes could examine how current and future water levels affects soil conditions, plant assemblage structure, and nutrient losses/storage over time. Such studies would provide useful feedback to help guide restoration of wetlands for water quality improvements and carbon storage.