Implementation of a Real-Time Forecasting System to Control Selenium Contamination in the Delta Mendota Canal (DMC)
* Is it possible to improve the quality of water in a canal with real-time monitoring and remote control of known contaminators according to assimilative capacity?
We predict that we can greatly improve water quality in the DMC and Mendota Pool and help to protect the wetlands of the Pacific flyway from degradation.
We will design and install a system that will allow us to use the internet to monitor and operate six sumps that discharge seleniferous water into the DMC. We will link this system to an existing hydrologic model to predict the quality of water in the canal.
Need and Benefit
The purpose of the DMC and related facilities is to deliver Central Valley Project (CVP) water to the water-deficient San Joaquin Valley. The DMC extends 116 miles from the Sacramento-San Joaquin Delta to the Mendota Pool on the San Joaquin River. Water conveyed in the northern 70 miles of the DMC is delivered to contractors for municipal and domestic supply, irrigation, stock watering, contact recreation, and wildlife habitat. Water conveyed in the lower 46 miles is delivered to contractors for irrigation, stock watering, warm freshwater habitat, and wildlife habitat. Wetlands in the region are important to birds migrating along the Pacific Coast of North America.
The California Regional Water Quality Control Board, with approval of the U.S. Environmental Protection Agency (EPA), has established a water quality objective of two parts per billion (ppb) monthly average concentration of selenium in water in the Grassland wetland supply channels. The DMC is the primary source of water for these channels.
The lower 46 miles of the canal are contaminated with selenium from inflows of surface and ground water. Reclamation has measured selenium concentrations in the DMC and Mendota Pool since 1985 under orders from the Regional Board. Currently, Reclamation collects daily composite samples of water from four places along the canal and weekly samples from the six sumps that discharge subsurface drain water into the canal. The sumps discharge less than two cubic feet per second (cfs) of water that has over 600 ppb of selenium. Analysis of the data collected by Reclamation reveal the following facts:
* Violation of the monthly average selenium objective of two ppb has occurred during winter and early spring when flow in the DMC is reduced. Low flow in the DMC provides little capacity for DMC to assimilate selenium.
* Selenium concentration in the DMC during the winter and early spring months is associated with rainfall events. Rainfall that infiltrates the soil helps to recharge local ground water levels. Rainfall-runoff accumulates downslope and often ponds immediately upstream of the DMC which acts like a dam impeding natural ground water flow towards the San Joaquin River. This truncating of the natural flow path causes water levels in the sumps to rise quickly and trigger the sump pumps when water levels rise above a set sensor height.
The six sumps are located about 65 miles from Fresno, California. The pump in each sump is operated with a float sensor that drains the sump into the canal when the depth exceeds a specified depth. During winter and early spring, the sumps discharge seleniferous water continuously into the empty DMC. We do not have field staff who can turn the sumps off. We have no way of knowing when the sumps are running or how contaminated their water is.
We predict that if we could operate the six sumps according to the assimilative capacity of the DMC, we can greatly improve water quality in the canal and Mendota Pool and help to protect the wetlands of the Pacific flyway from degradation.
In 2004, the S&T Program generously funded the development of a Real-Time Forecasting and Emergency Management System computer model of the canal and the sumps for real-time monitoring and forecasting of selenium in the DMC (Project #947, Real-time monitoring, modeling and management of selenium drainage discharges in the Delta Mendota Canal, 2004). We worked with the Danish Hydrologic Institute (DHI) to create the model using MIKE 11 software. This model is designed to use local weather data, real-time flow data, and _in situ_ electrical conductivity sensors to predict water quality conditions in the sumps and canal.
The proposed supervisory control and data acquisition (SCADA) system is needed to monitor and control the flow and quality of water in each sump and in the canal.
The benefit will be a significant improvement in the quality of water in the DMC, Mendota Pool, and grassland wetland
Contact the Principal Investigator for information about these documents.