Zhao, Z., Ashraf, M. I. and Meng, F.-R. 2013. Model prediction of soil drainage classes over a large area using a limited number of field samples: A case study in the province of Nova Scotia, Canada. Can. J. Soil Sci. 93: 73-83. Soil drainage maps are frequently required for crop, forest, and environmental management. However, modelling soil drainage over a large area (>1000 km2) is difficult due to complex soil-forming processes, large spatial variations, and the limited number of field samples, which are often insufficient to reflect local variations. In this study, a two-stage approach was used to produce soil drainage maps over a large area (the province of Nova Scotia). In the first stage, an existing soil drainage model developed in a small watershed with a sufficient number of field samples that could represent local topography was adopted. As a comparison, an artificial neural network model was built and calibrated with 1545 field samples across the province of Nova Scotia. Both models were used directly to predict soil drainage maps in the province of Nova Scotia. Results indicate that both models produced poor predictions. In the second stage, after dividing the entire provincial area into sub-areas (landforms) based on different division methods, corresponding linear transformation models were subsequently developed to adapt soil drainage classes produced by a base model (the existing soil drainage model) to fit field samples. Parameters of linear transformation models were estimated with field samples. Results indicate that the best linear transformation model was composed of 12 linear equations corresponding to 12 landforms (combinations of ecoregion and texture), and improved the prediction of rapidly drained (9.6%), well-drained (21.3%), moderately well-drained (14.1%), and imperfectly drained (7.5%) plots compared with the base model. Thus, the two-stage approach can obviously improve the accuracy of predicted soil drainage classes over a large area.
You have requested a machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Neither BioOne nor the owners and publishers of the content make, and they explicitly disclaim, any express or implied representations or warranties of any kind, including, without limitation, representations and warranties as to the functionality of the translation feature or the accuracy or completeness of the translations.
Translations are not retained in our system. Your use of this feature and the translations is subject to all use restrictions contained in the Terms and Conditions of Use of the BioOne website.
Vol. 93 • No. 1