No abstract

Land evaluation is a critical step in land-use planning. Although many methods have been developed since the formulation of the FAO framework for land evaluation, several of the more traditional approaches still remain in widespread use but have not been adequately evaluated. Contrary to more recent land evaluation systems, which need considerable data, these systems only require basic soil and landscape information to provide a general view of land suitability for major types of land use. As the FAO initially presented its qualitative framework for land-use planning, based on two previous methods developed in Iran and Brazil, in this study we assessed the reliability and accuracy of a traditional land evaluation method used in Iran, called land classification for irrigation (LCI), comparing its results with several qualitative and quantitative methods and actual yield values. The results showed that, although simpler than more recently developed methods, LCI provided reliable land suitability classes and also showed good relationships both with other methods analysed and with actual yields. Comparisons between qualitative and quantitative methods produced similar results for common crops (a barley-alfalfa-wheat-fallow rotation). However, these methods performed differently for opportunist crops (such as alfalfa) that are more dependent on income and market conditions than on land characteristics. In this work, we also suggest that using the FAO method to indicate LCI subclasses could help users or managers to recognize limitations for land-use planning.

25Detailed soil surveys involve costly and time-consuming work and require expert knowledge. Since soil surveys provide information to meet a wide range of needs, new methods are necessary to map soils quickly and accurately. In this study, multilayer perceptron artificial neural networks (ANN) were developed to map soil units using digital elevation model (DEM) attributes. Several optimal ANNs were produced based on a number of input data and hidden units. The approach used 30 test and validation areas to calculate accuracies of interpolated and extrapolated data. The results show that the system and level of soil classification employed had a direct effect on the accuracy of the results. At the lowest level, lower errors were observed with the WRB classification criteria than the Soil Taxonomy (ST) system, but more soil classes could be predicted when using ST (seven soils in the case of ST vs. five with WRB). Training errors were below 11% for all the ANN models 35applied, while the test (interpolation error) and validation (extrapolation error) errors were as high as 50% and 70%, respectively. As expected, soil prediction using a higher level of classification presented a better overall level of accuracy. To obtain better predictions, in addition to DEM attributes, data related to landforms and/or lithology, as soil-forming factors, should be used as ANN input data. 40

No abstract

No abstract