Effects of Sedimentation on Phosphorus Retention in Seasonally Submerged Wetland Soils.
Jonathan Maynard1, Anthony O'Geen1, Jaiyou Deng1, Neil Brauer2, Denise Tu2, and Randy Dahlgren1. (1) Univ of California, Davis, One Shields Ave., Davis, CA 95616, (2) Univ of California, Davis, One Shields Ave., Davis, CA 95616
Irrigation induced erosion is one of the leading sources of pollution affecting water quality nationwide. In California's Central Valley, constructed wetlands (CWs) are used as best management practices for the treatment of agricultural return flows that contain high levels of sediment and nutrients. These wetlands are designed to treat agricultural drainage water before it is re-circulated into the San Joaquin River. Retention of suspended solids is one of the primary treatment mechanisms of constructed wetlands. Sedimentation and burial is also an important mechanism of phosphorus (P) removal from the water column because suspended particles sorb P. Sorption capacity of wetlands can be diminished in instances where suspended sediment is low and P loads are high, as sorption sites become saturated and are not replenished by new material.The retention of dissolved-P is highly correlated to the concentration of extractable amorphous iron. Wetland soils with high sesquioxide contents have higher P retention capacities. This suggests that dissolved-P retention is dependant on soil mineralogy and pedogenic processes that form sesquioxides. Sedimentation also promotes the burial of both sorbed and organicly bound phosphorus. The objectives of this study are to examine the relationships between sedimentation patterns, the development of poorly crystalline Fe, and the fate of P in CWs.We examined the spatial relationships between wetland soil properties and their effects on the bio-availability of phosphorus. A 13 year-old constructed wetland receiving agricultural tailwaters was monitored during the 2004 and 2005 irrigation seasons (May-Sept.). Weekly water samples were taken at the input and output and analyzed for total suspended solids and total phosphorus. Soil samples were collected at the start of year 1 and sedimentation samples were collected at the end of years 1 and 2. Samples were collected throughout the wetland (n=50) and analyzed for organic carbon, total nitrogen, total phosphorus, Olson-P, particle size, soil organic matter (SOM), crystalline and poorly crystalline iron- and manganese-oxides, and phosphorus sorption index (PSI). A digital terrain model of the wetland floor was created to examine the spatial relationships between soil topography, hydraulic flowpath, sedimentation, and phosphorus retention. Water quality monitoring results indicated that CWs removed 97% of the suspended sediment from tail water. The average P removal efficiency was 71%. Sediment accumulation was as high as 68 kg m-2 y-1 in areas proximal to the input location. Deposition was much lower in areas distal to the input, where sediment accumulation rates were as low as 0.5 kg m-2 y-1. Selective dissolution was used to measure crystalline and poorly crystalline iron-oxides. Total Fe (Fed) ranged from 6 g kg-1 in the low depositional zones to 18 g kg-1 in high depositional zones. Poorly crystalline Fe (Feo) ranged from 3.0 g kg-1 in the high depositional zones to 7.3 g kg-1 in low depositional zones. Feo /Fed ratios were highest in the low depositional zones (0.73) and lowest in the high depositional zones (0.22) suggesting that poorly crystalline Fe is formed in specific pedogenic environments within the CW. Geostatistical interpolation maps were created to analyze the spatial relationships between soil properties. Randomization tests (Partial Mantel Tests) were used to determine the spatial variability of each soil property and to determine which properties best explained the variability of both PSI and bio-available P (BAP) after accounting for the effects of spatial autocorrelation. PSI had the highest partial correlation with clay sized particles( r=0.403 p=0.000) and poorly crystalline Fe (r=0.369 p=0.000). BAP had a high negative partial correlation with Feo (r=-0.960 p=0.022) and high partial correlation with clay (r=0.460 p=.000). BAP is low where poorly crystalline Fe is high, due to its high sorption capacity. BAP is high were clay content is high due to the weak associations of P with clay edges. Moreover, autocorrelation in spatial patterns demonstrated that concentrations of Feo are lowest at the inlet, increasing with distance along the flowpath. Flowpath plays a major role in the spatial distribution of sedimentation, and thus the bio-availability of phosphorus. The formation of poorly crystalline iron oxides appears to be controlled by sedimentation rates. Furthermore, systematic patterns in pedogenesis and associated formation of poorly crystalline iron compounds, plays an important role in the degree of P retention in these seasonally submerged soils.