Suspended sediment and total phosphorus loadings from small agricultural watersheds in western Illinois

Russell, Amy M.

Permalink

https://hdl.handle.net/2142/46697 Copy

Description

Title

Suspended sediment and total phosphorus loadings from small agricultural watersheds in western Illinois

Author(s)

Russell, Amy M.

Issue Date

2014-01-16T17:59:31Z

Director of Research (if dissertation) or Advisor (if thesis)

Cooke, Richard A.

Department of Study

Engineering Administration

Discipline

Agricultural & Biological Engr

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Keyword(s)

• load calculation
• regression
• sediment
• phosphorus

Abstract

On larger rivers, instantaneous sample concentrations are often treated as being representative of the mean daily concentration for load calculation purposes. This assumption, however, is not appropriate on streams where concentrations can change substantially within a day. In five small, rural watersheds in western Illinois, the collection and analysis of data during runoff events was done on a sub-daily time step. To have accurate load estimates, the selected load calculation method should correctly characterize loading behavior during short duration runoff events. The use of statistical models with residuals-based error correction (i.e. the composite method) has become an increasingly popular technique for load calculations. This study is an application of error corrected regression models to compute continuous records of suspended sediment and total phosphorus concentrations at five watersheds in western Illinois. Due to the small drainage areas of the studied streams, all regression models were developed and applied using a 15-minute time-step. Four methods of constructing continuous concentration records were compared, and the best method to compute sediment and phosphorus loads for a 10-year period of study was identified. For both suspended sediment and total phosphorus, load calculations by error corrected regression models produced estimates that were the most precise and least biased. Further, the method of error correction was not as critical as the act of error correction itself. During the ten-year study period, 5% of the record accounted for approximately 50% of the flow, 91% of the total phosphorus load and more than 96% of the sediment load. On average, the 1-day maximum accounted for 10% of annual flow and more than 30% of annual sediment and phosphorus loads.

Graduation Semester

2013-12

Permalink

http://hdl.handle.net/2142/46697

Copyright and License Information

Copyright 2013 Amy Russell

Owning Collections