Skip to main content
ARS Home » Research » Publications at this Location » Publication #260535

Title: Spatial considerations in wet and dry periods for phosphorus in streams of the Fort Cobb watershed, USA

Author
item Franklin, Dorcas
item Steiner, Jean
item Duke, Sara
item Moriasi, Daniel
item Starks, Patrick

Submitted to: Journal of the American Water Resources Association
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/25/2013
Publication Date: 4/3/2013
Citation: Franklin, D.H., Steiner, J.L., Duke, S.E., Moriasi, D.N., Starks, P.J. 2013. Spatial considerations in wet and dry periods for phosphorus in streams of the Fort Cobb watershed, USA. Journal of the American Water Resources Association. 49(4):908-922. DOI: 10.111/jawr.12048.

Interpretive Summary: While it is critical to include phosphorus in management programs designed to decrease eutrophication of freshwaters, it is still unclear to what extent the landscape and associated management influence nutrient concentrations in streams. We theorized that the influence of landscape characteristics will vary between regions. Therefore, an understanding of background variations in surface-water phosphorus concentrations is necessary in the development of reliable predictive models. The Fort Cobb Watershed in Oklahoma (USA) has diverse landscape characteristics and provides an opportunity to explore water quality within a diverse landscape during both a Wet Phase (04/2007-12/2009) and a Dry Phase (01/2005-03/2007). Scientists from USDA ARS Grazinglands Research Laboratory and J. Phil Campbell Sr., Natural Research Conservation Center conducted research to identify patterns associated to location (spatial dependence) and time for dissolved P and relate those to landscape characteristics. From 2005-2009, bi-weekly 1-L samples were collected at 15 sites distributed throughout the stream network and analyzed for dissolved phosphorus. Landscape characteristics were compiled for each stream basin, within each stream reach and from climate data and related to dissolved P concentrations. Results indicated that spatial dependence was present and there was significantly more spatial dependence during the Wet Phase than during the Dry Phase. Dissolved P stream concentrations were 3 to 5 times higher in the Wet Phase than the Dry Phase. Additional analysis indicated that landscape characteristics (topography, soil and geology) were better predictors for dissolved P when there was spatial dependence. During Wet Phase, landscape characteristics which identified rapid water movement were associated with higher dissolved P. This research demonstrated that varied background dissolved P concentrations can be a result of both landscape and climatic influences. In addition, we demonstrated that we can discern with some certainty the areas of a watershed that may be more vulnerable to phosphorus losses and/or contaminations during climatic swings.

Technical Abstract: While it is critical to include P in management programs designed to decrease eutrophication of freshwaters, it is still unclear to what extent the landscape and associated management influence nutrient concentrations in streams. We theorized that the influence of a landscape metric will vary between ecoregions or biogeophysical settings. Therefore, an understanding of background variations in surface-water phosphorus concentrations is necessary in the development of reliable predictive models. The Fort Cobb Watershed in Oklahoma (USA) has diverse biogeophysical settings and provides an opportunity to explore the association of water quality with a diverse group of landscape metrics during both a Wet Phase (04/2007-12/2009) and a Dry Phase (01/2005-03/2007). The objective of this work is to identify spatial and temporal patterns in dissolved P [soluble reactive P (SRP) and bioavailable P (BAP)] and relate those to landscape metrics. From 2005-2009, bi-weekly 1 L samples were collected at 15 sites distributed throughout the stream network and analyzed for SRP and BAP. During statistical analysis, spatial autocorrelation (SAC) was evaluated using contiguity and upstream connectivity matrices. Horizontal, longitudinal and vertical biogeophysical metrics were compiled for each contributing area, within each stream reach and from climate data, respectively and related to dissolved P concentrations. Results for both SRP and BAP indicated that spatial dependence was present (p < 0.05) for the Contiguity and the Upstream matrix and there was significantly more SAC in the Wet Phase than in the Dry Phase. Dissolved P stream concentrations were 3 to 5 times higher in the Wet Phase than the Dry Phase (p < 0.05). Analysis with recursive partitioning, by Dry and Wet Phase and presence or absence of SAC resulted in higher R2 when SAC was present than without SAC and indicated that Horizontal Variables (topography, soil, geology, management) were better predictors for SRP when there was SAC. During Wet Phase, Horizontal Variables which identified rapid water movement were associated with higher dissolved P. Varied background dissolved P stream concentrations can be as a result of both geophysical and climatic influences. We can discern with some certainty areas vulnerable to P losses and/or contaminations due to either drought or very wet conditions.