Pedotransfers Functions

In this study, the performance of 63 existing pedotransfer functions (PTFs) is evaluated to estimate oven-dry soil bulk density (BD) by using a dataset of 3316 soil cores taken in the farmlands of Campania (southern Italy). As expected, the lack of direct calibration yields prediction accuracy from unsatisfactory to rather weak. Therefore, we advance the working hypothesis that the use of hierarchical soil mapping information can make the application of existing PTFs more reliable. We show that grouping data according to land-systems classes or soil groups considerably improves the prediction ability quantified through the root mean squared error (RMSE) and the coefficient of determination (R 2). An independent data set of 105 soil cores taken from two hillslopes in the Upper Alento River Catchment in southern Campania was used to verify our assumption. The validation step shows that the knowledge of a soil-landscape map is an efficient tool for improving the prediction of BD. This approach will be employed in a subsequent study to develop site-specific PTFs for the study region.

Soil bulk density (ρ b) data are needed for a wide range of environmental studies. However, ρ b is rarely reported in soil surveys. An alternative to obtain ρ b for data-scarce regions, such as the Rio Doce basin in southeastern Brazil, is indirect estimation from less costly covari-ates using pedotransfer functions (PTF). This study primarily aims to develop region-specific PTFs for ρ b using multiple linear regressions (MLR) and random forests (RF). Secondly, it assessed the accuracy of PTFs for data grouped into soil horizons and soil classes. For that purpose, we compared the performance of PTFs compiled from the literature with those developed here. Two groups of data were evaluated as covariates: 1) readily available soil properties and 2) maps derived from a digital elevation model and MODIS satellite imagery, jointly with lithological and pedological maps. The MLR model was applied step-wise to select significant predictors and its accuracy assessed by means of cross-validation. The PTFs developed using all data estimated ρ b from soil properties by MLR and RF, with R 2 of 0.41 and 0.51, respectively. Alternatively, using environmental covariates, RF predicted ρ b with R 2 of 0.41. Grouping criteria did not lead to a significant increase in the estimates of ρ b. The accuracy of the 'regional' PTFs developed for this study was greater than that found with the 'compiled' PTFs. The best PTF will be firstly used to assess soil carbon stocks and changes in the Rio Doce basin.

Soil bulk density is a key property in defining soil characteristics. It describes the packing structure of the soil and is also essential for the measurement of soil carbon stock and nutrient assessment. In many older surveys this property was neglected and in many modern surveys this property is omitted due to cost both in laboratory and labour and in cases where the core method cannot be applied. To overcome these oversights pedotransfer functions are applied using other known soil properties to estimate bulk density. Pedotransfer functions have been derived from large international data sets across many studies, with their own inherent biases, many ignoring horizonation and depth variances. Initially pedotransfer functions from the literature were used to predict different horizon type bulk densities using local known bulk density data sets. Then the best performing of the pedotransfer functions were selected to recalibrate and then were validated again using the known data. The predicted coefficient of determination was 0.5 or greater in 12 of the 17 horizon types studied. These new equations allowed gap filling where bulk density data were missing in part or whole soil profiles. This then allowed the development of an indicative soil bulk density map for Ireland at 0–30 and 30–50 cm horizon depths. In general the horizons with the largest known data sets had the best predictions, using the recalibrated and validated pedotransfer functions.

Soil water retention is a key property for the study of water available for plants, infiltration, drainage, hydraulic conductivity, irrigation, plant water stress and solute movement. However, its measurement is difficult, expensive and time consuming. An alternative is the use or development of "pedotransfer functions (PTFs)" as models to predict these properties from readily available data. Due to the few amount of soil data in the Caramacate Riverupper basin of Venezuela,widely used functions developed in other locationswere assessedin order to estimate the soil water contentat-33kPa (WR-33kPa). The statistical measures to assess the performanceof the PTFs were the mean error (ME), the root mean square error (RMSE) and the coefficient of determination. The performance of all evaluated PTFs was compared with an equation generated for the study area. Results show that the Urach´s PTF was to estimate WR-33 kPa in the soils of the study area. The weak performance of the other PTFs could be related to the fact that these mountain soils are rich in 2:1 clay (vermiculite) and smectite and high stoniness, which were not used as independent variables of those PTFs to estimate the WR-33 kPa. Therefore, it could be convenient to generate equations that fit the characteristics of the soils of the study area, with even a small number of samples.