Title of article :
Spatial patterns of, and environmental controls on, soil properties at a riparian–paddock interface
Author/Authors :
Smith، نويسنده , , M. and Conte، نويسنده , , P. and Berns، نويسنده , , A.E. and Thomson، نويسنده , , J.R. and Cavagnaro، نويسنده , , T.R.، نويسنده ,
Pages :
8
From page :
38
To page :
45
Abstract :
Riparian zones are prominent features of agricultural landscapes because they are the last point to intercept nutrients and sediments before they enter water bodies. We investigated the soil properties, nutrient dynamics and vegetation composition at the riparian–agriculture interface. Soil physicochemical and vegetation properties were spatially heterogeneous along the transition from the grazed paddock into the un-grazed and revegetated riparian zone. Soil C stocks varied considerably across the site, with values ranging from 2% in the paddock to 5% in the riparian zone. Using Bayesian model selection, a predictive model for total soil carbon was developed. By including soil moisture content and canopy cover in the model, it was possible to predict total soil carbon with 80% accuracy at the site level and 87% at the transect level. This opens up the potential for total soil carbon levels to be estimated by the quantification of easily measured ecosystem properties. Analysis of the chemical nature of the carbon in theses soils by solid state 13C NMR spectroscopy, showed the presence of more recalcitrant forms of carbon in the revegetated riparian zone compared to the grazed paddock. Spatial patterns of soil mineral N pools were highly variable ( NO 4 + − N ranged from 1 to 5 μg/g dry soil; NO 3 − − N ranged from 0.4 to 2.2 μg/g dry soil); however, clear patterns in potentially mineralizable N (PMN) were observed, with rates of PMN in the paddock being less than half of those adjacent to the stream in the riparian zone. Results are discussed in the context of the dynamic nature of soil processes at the agriculture – riparian interface, and the potential to develop models to predict soil carbon using easily measurable vegetation and soil properties.
Keywords :
Bayesian modelling , Nuclear magnetic resonance spectroscopy (NMR) , Riparian restoration , Soil carbon , soil nutrients , Soil respiration , nitrogen cycling
Journal title :
Astroparticle Physics
Record number :
1999628
Link To Document :
بازگشت