Predictors of Nitrogen-Fixing Activity Across a Local Gradient in Fire History for a Temperate Semiarid Grassland.
Poster Number 24
See more from this Division: PosterSee more from this Session: Soil Economics 101: Evaluation of Ecosystem Services
We sampled 18 plots from a mixed-grass prairie site in the Northern Great Plains. Plots had variable prescribed fire histories (e.g. time since fire) and were sampled five times from September 2011 - August 2012. For each, we quantified 21 soil properties and the activity of the nifH gene (DNA and mRNA) using quantitative PCR. Multiple linear regressions were performed to determine the best predictor(s) of variation in nifH mRNA copy number. Predictor variables included 21 soil properties, two weather variables, and two fire history variables.
The seasonal sampling tracked the progression of a regional drought. Greater levels of nifH DNA and mRNA were observed in early (moist) than late (drought stage) sampling. For the moist period, we determined the greatest amount of variation in nifH mRNA was explained by a model with 4 variables (R2= 0.35). The nifH mRNA was positively correlated with manganese and negatively correlated with iron, sulfur, and temperature. Based on comparisons of squared semi-partial coefficients of determination, we determined three variables were equally influential (temperature, iron, and sulfur) followed by manganese.
Soil moisture was a main factor limiting nifH activity. Some soil properties were also useful predictors of nifH mRNA. However, we failed to detect correlations between nifH activity and either fire history variables, nitrogen, or phosphorus. A small number of other soil properties, exhibiting patterns of collinearity [e.g. manganese correlated with sulfur (r= 0.42) and iron (r=0.80)], were useful predictors of variation in nifH activity.
See more from this Session: Soil Economics 101: Evaluation of Ecosystem Services