English

Abstract

Water quality in the lower Colorado and Gunnison Rivers in western Colorado, and many of their tributaries, is impaired by selenium, which originates from the local Mancos shale. Because of the diffuse and widespread nature of this source, there are limited opportunities to reduce selenium inputs. One option is to treat selenium-contaminated surface water at strategic locations, such as point-source discharges. Gravel extraction is common along these rivers, and treatment of discharges from pit dewatering presents an opportunity for reducing selenium loading. The goal of this study is to determine: 1) whether a passive selenium-reducing bioreactor can accomplish high-efficiency selenium removal from the basic, saline water typical of the Grand Valley; 2) whether zero-valent iron is beneficial as a bioreactor component; and 3) optimum detention time. To date, bacterial reduction of selenium has been successfully accomplished using power-, and equipment-intensive active treatment systems. The passive bioreactors we are testing can function unattended, ideally by a gravity feed (no pumps), and the fuel for the bacteria would be agricultural wastes (e.g., wood chips, hay, cow manure inoculum) and other materials (e.g., quarried limestone) collected locally. Four 208-liter (55 gallon) bioreactors were constructed with varying amounts of cow manure, hay, sawdust, wood chips, limestone, and zero-valent iron. Influent to the reactors is drawn from a dewatering trench in a gravel pit next to the Colorado River near Grand Junction, Colorado. The reactors were operated, with varying detention times, over a six-month period from July through November 2006. The results of this study demonstrate that passive bioreactors can accomplish up to 98% removal of Se from surface and ground waters in the Grand Valley of western Colorado. A bioreactor designed to promote microbial processes functioned as efficiently as reactors incorporating ZVI, in spite of the potential of the ZVI to enhance the biological removal process. The highest removal rates were achieved using a detention time of 12 hours, but circumstances prevented optimization of detention time. Additional Keywords: Selenium reduction; passive treatment; agricultural runoff