Effects of the herbicide glyphosate on nitrogen cycling in an acid forest soil

Abstract

The herbicide glyphosate was sprayed aerially on a section of conifer forest in Atlantic Canada that had been previously clearcut and reforested. Glyphosate was then tested for effects on ammonification, nitrification, and denitrification for a period of 8 months by comparing microbial activity in treated and untreated zones of the clay loam forest soil and the overlying decomposing litter, both with a pH of 3.8. With ammonification, there was generally a stimulation of activity in both the forest litter (FL) and forest soil (FS) that had been exposed to glyphosate during spraying. Nitrification rates in FL and FS were very low and glyphosate had no appreciable stimulatory or inhibitory effect on nitrification. Although glyphosate stimulated denitrification in a few instances, it generally had no significant effect on denitrification activity in FL and FS exposed during spraying. With all processes, microbial activity in FL was significantly greater than that in FS. Laboratory bioassays were also performed with FL and FS, as well as two silt loam (pH 5.8 and 6.4) and one sandy loam (pH 6.8) agricultural soils, using glyphosate concentrations up to 200 times higher than field application rates. With ammonification and denitrification, glyphosate generally stimulated activity at all levels tested and in all soil used. Glyphosate stimulated ammonification by 50% at concentrations ranging from 140 to 550 μg g-1 for the soils and >4000 μg g-1 for FL. With denitrification, the corresponding herbicide levels were approximately 2250 μg g-1 for FS, > 10,000 for FL, and 450 for an agricultural soil. With nitrification, it was estimated that glyphosate concentrations greater than 1000 to 2000 μg g-1 would be required to cause a 50% inhibition of activity. The careful use of glyphosate in forestry should have no toxic effects on N cycling in soils. © 1991 Kluwer Academic Publishers.