Examining the Fate of Metal Oxide Nanoparticles in Water Saturated Soil Environments

This paper applies geostatistics procedures to model soil test phosphorus (STP) in 11km Castor Watershed in Southern Quebec, Canada. Phosphorus in the soil is impacted by processes such as landuse pattern, management practices, soil properties, geomorphology etc. This is a challenge when developing or determining the covariance functions for the experimental variogram in geostatistical modeling. Ordinary kriging (OK) method was used to estimate the STP at unsampled points, but this has a smoothing effect. This smoothing underestimate high values and overestimate low values. Therefore, there is need for an efficient technique such as the sequential Gaussian simulation (SGS) which was utilized to draw from the conditional probability distribution at the simulation grid nodes. SGS produced several equal-probable realizations. SGS reproduced the original data better compared to OK where several of the high values were masked during estimation. This result is very relevant especially in the reduction of phosphorus (P) load into Missisquoi Bay, southern Quebec, Canada where there is reported cases of eutrophication due to excess P from non-point sources. In addition, the results could be very relevant in nutrient use efficiency, classification of management zones in form of polygons, adaptive agricultural management practices and optimization of Phosphorus-Index for land-use and ecosystem management.