Predict Bulk Density Research Articles

Development of a fluid bed granulation process control strategy based on real-time process and product measurements

This article describes the results of three case studies conducted consecutively, in order to develop a process control strategy for a top-spray fluid bed granulation process. The use of several real-time particle size (i.e., spatial filter velocimetry and focused beam reflectance measurement) and moisture (i.e., near infrared (NIR) and Lighthouse near infrared spectroscopy) analyzers was examined. A feed-forward process control method was developed, where in-line collected granulation information during the process spraying phase was used to determine the optimum drying temperature of the consecutive drying phase. Via real-time monitoring of process (i.e., spraying temperature and spray rate) and product (i.e., granule size distribution and moisture) parameters during the spraying period, the batch bulk density was predicted at the end of the spraying cycle, using a PLS model. When this predicted bulk density was not meeting the desired value, the developed control method allowed the calculation of an adjusted drying temperature leading to the desired batch bulk density at the end of the granulation process. Besides the development of the feed-forward control strategy, a quantitative PLS model for in-line moisture content prediction of the granulated end product was built using the NIR data.

Estimation of bulk density of waterlogged soils from basic properties

Pedotransfer functions (PTFs) to predict bulk density (BD) from basic soil data are presented. Available data pertaining to seasonally impounded shrink–swell soils of Jabalpur district in the Madhya Pradesh state of India were used for the study. The data included horizon-wise information of 41 soil profiles in the study area covering nearly 5 million ha. Six independent variables, namely textural data (sand, silt and clay), field capacity (FC), permanent wilting point (PWP) and organic carbon content (OC) were used as input in hierarchical steps to establish dependencies, with bulk density as the dependent variable, using statistical regression and artificial neural networks. The PTFs derived using neural networks [average root mean square error (RMSE) 0.05] were relatively better than statistical regression PTFs (average RMSE > 0.1). The best-performing PTFs required input data on sand, silt content, FC and PWP, with lowest prediction errors (RMSE 0.01, maximum absolute error (MAE) 0.01) and highest values of index of agreement (d, 0.95) and R 2 (0.65). Use of measures of structure, as well as information on pore structure, was found to be essential to derive acceptable PTFs. Inclusion of OC as an input variable showed relatively better fitting to the training data set, implying an underlying relationship between OC and BD, but the neural networks could not mimic the relationship when tested against subset.

Empirically‐derived pedotransfer functions for predicting bulk density in European soils

The lack of comprehensive data on the bulk density of soil types at the European scale is a serious limitation for pan‐European environmental risk assessment studies. Although many predictive methods have been published, most have limitations for application across Europe. We therefore developed a semi‐empirical method of prediction using a large UK dataset and tested it and some other methods against a pan‐European dataset. Our method indicated that five separate conceptual groupings of the development dataset were valid. Predictive equations based on multiple regression analysis for each of the five groups explained between 40 and 69% of the measured variation in each one. When used to predict measured bulk density from the European dataset, the equations explained 63% of the measured variation in mineral horizons from soil environments similar to those of the development dataset with a predictive mean percentage error of ±11%. The equation for organic horizons explained 29% of the measured variation in bulk density with a mean percentage error of ±39%. For those horizons from soil environments outside those of the development dataset, prediction of bulk density was relatively poor, even when using soil region‐specific PTFs derived from its data. It was concluded that, for these soils, factors other than organic carbon, particle size, horizon depth, mechanical cultivation or parent material have a major influence on bulk density and need further investigation.

The prediction of the bulk densities of powder mixtures, and its relationship to the filling of hard gelatin capsules.

From measurements of the properties of individual components, it has been found possible to predict the maximum tapped bulk density of two component mixtures of a range of particle size fractions of acetylsalicylic acid and lactose in varying proportions. This has been extended to the prediction of bulk density of such mixtures when filled into hard gelatin capsules by a system which results in the powders existing at the maximum tapped bulk density. Consequently the capsule fill weight can also be predicted for such systems. The method is less satisfactory for the prediction of the bulk density and capsule fill weight, when the capsules are filled by a process involving compression of the powder within the capsule shell.

Prediction of Bulk Density and Molecular Packing in Model Dendrimers with Different Chain Stiffness

By means of molecular dynamics simulations of model dendrimers, we analyze the dependence of the bulk density and molecular packing on the dendrimer molecular weight and intrinsic stiffness. We find that the density is consistently higher in flexible dendrimers than in the rigid ones with a large bending angle. The density values change slightly within the first two generations to reach a plateau. We interpret these results in terms of free volume, showing that the enhanced accessible free volume that characterizes the end-dendron monomers is counterbalanced by the higher number of internal monomers, leading to a constant bulk density for generations larger than three. The added stiffness affects the geometrical properties and the molecular rearrangement of the bulk, reducing the short-range local order and the packing efficiency favoring the dendrimer interpenetration. Our prediction for the bulk density matches and rationalizes experimental and previous all-atom simulation results.

A fundamental approach and its experimental validation to simulate density as a function of moisture content during drying processes

Although several empirical models are available in the literature to predict density in solid matrices, only a very limited number of theoretical models have been reported. So far, no model considered the possible variation in the initial air volume existing at the beginning of the drying process. In this contribution, a theoretical model to predict bulk density of dried materials was built by considering two mechanisms that might occur during drying processes. These mechanisms are represented by collapse and shrinkage functions. The predictions obtained by this theoretical model were extensively validated with experimental data published by several independent groups for different food products dried with different technologies. In all these cases, the model gave excellent agreement with the experimental data regardless the topology of the curve bulk density versus moisture content. The model was also compared with other published models. The result of this comparison revealed that the errors resulting from the predictions obtained by the present model are among the smallest. Shrinkage and collapse functions were used to analyze the mechanisms by which bulk density varies during air-drying and freeze-drying. The model showed that both shrinkage and collapse phenomena are dramatically involved during air-drying. However, in the case of freeze-drying, no collapse is observed and only partial shrinkage is taking place. Hence, the present model can be used as a tool to predict the bulk density with excellent accuracy, to understand the dynamic mechanisms involved during drying. Moreover, this model can be incorporated to other models involving the variation of density as a function of moisture .

Near infrared spectroscopy for soil bulk density assessment

Soil bulk density values are needed to convert organic carbon content to mass of organic carbon per unit area. However, field sampling and measurement of soil bulk density are labour‐intensive, costly and tedious. Near‐infrared reflectance spectroscopy (NIRS) is a physically non‐destructive, rapid, reproducible and low‐cost method that characterizes materials according to their reflectance in the near‐infrared spectral region. The aim of this paper was to investigate the ability of NIRS to predict soil bulk density and to compare its performance with published pedotransfer functions. The study was carried out on a dataset of 1184 soil samples originating from a reforestation area in the Brazilian Amazon basin, and conventional soil bulk density values were obtained with metallic “core cylinders”. The results indicate that the modified partial least squares regression used on spectral data is an alternative method for soil bulk density predictions to the published pedotransfer functions tested in this study. The NIRS method presented the closest‐to‐zero accuracy error (−0.002 g cm−3) and the lowest prediction error (0.13 g cm−3) and the coefficient of variation of the validation sets ranged from 8.1 to 8.9% of the mean reference values. Nevertheless, further research is required to assess the limits and specificities of the NIRS method, but it may have advantages for soil bulk density predictions, especially in environments such as the Amazon forest.

Pyrazine in Supercritical Xenon: Local Number Density Defined by Experiment and Calculation

Toward our goal of using supercritical fluids to study solvent effects in physical and chemical phenomena, we develop a method to spatially define the solvent local number density at the solute in the highly compressible regime of a supercritical fluid. Experimentally, the red shift of the pyrazine n-pi* electronic transition was measured at high dilution in supercritical xenon as a function of pressure from 0 to approximately 24 MPa at two temperatures: one (293.2 K) close to the critical temperature and the other (333.2 K) remote. Computationally, several representative stationary points were located on the potential surfaces for pyrazine and 1, 2, 3, and 4 xenons at the MP2/6-311++G(d,p)/aug-cc-pVTZ-PP level. The vertical n-pi* ((1)B(3u)) transition energies were computed for these geometries using a TDDFT/B3LYP/DGDZVP method. The combination of experiment and quantum chemical computation allows prediction of supercritical xenon bulk densities at which the pyrazine primary solvation shell contains an average of 1, 2, 3, and 4 xenon molecules. These density predictions were achieved by graphical superposition of calculated shifts on the experimental shift versus density curves for 293.2 and 333.2 K. Predicted bulk densities are 0.50, 0.91, 1.85, and 2.50 g cm(-3) for average pyrazine primary solvation shell occupancy by 1, 2, 3, and 4 xenons at 293.2 K. Predicted bulk densities are 0.65, 1.20, 1.85, and 2.50 g cm(-3) for average pyrazine primary solvation shell occupancy by 1, 2, 3, and 4 xenons at 333.2 K. These predictions were evaluated with classical Lennard-Jones molecular dynamics simulations designed to replicate experimental conditions at the two temperatures. The average xenon number within 5.0 A of the pyrazine center-of-mass at the predicted densities is 1.3, 2.1, 3.0, and 4.0 at both simulation temperatures. Our three-component method-absorbance measurement, quantum chemical prediction, and evaluation of prediction with classical molecular dynamics simulation-therefore has a high degree of internal consistency for a system in which the intermolecular interactions are dominated by dispersion forces.

Study on drying and combustion process in grate-CFB incinerator

The drying and combustion process in the combined grate and circulating fluidized bed (grate-CFB) municipal solid waste (MSW) incinerator was investigated experimentally and mathematically. The drying grate bed was simulated by a muffle furnace, which could be controlled at a constant temperature level. The kind of wastes, thickness of waste layer fed and temperature were chosen as the adjustable parameters to study their effect on the drying process. The experimental results indicated that the hydrophilic wastes were more difficult to be dried than the hydrophobic wastes. The higher the temperature is the easier the waste is to be dried. The thinner waste layer is favorable to drying the waste. The pyrolysis experiment in the furnace showed that the higher temperature level could reduce the conversion rate of carbon to carbon monoxide. The semi-empirical mathematical model that included the bed material distribution submodel, volatile matter release submodel, carbon particle combustion submodel and so on was proposed. A 260 t/d grate-CFB incinerator was modeled and the model predicted bulk density agreed with the measured value from industrial field test. The predicted flue gas (e.g. CO2, CO) concentration deviated slightly from the industrial test data. The parameters and variables used in the model were determined by the experiments or practical field test. This model can be used to design the grate-CFB incinerator and guide its operation.

Comparing Spectral Soil Inference Systems and Mid‐Infrared Spectroscopic Predictions of Soil Moisture Retention

Mid‐infrared spectroscopy has been proposed as a cheap yet accurate alternative to a number of laboratory methods for measuring soil properties. While accurate predictions of a number of basic soil constituents have been reported, properties associated with soil structure have received far less attention. In this study, we looked at the efficacy of mid‐infrared reflectance spectroscopy in predicting moisture retention and whether better predictions can be achieved using pedotransfer functions using spectroscopic predictions of basic soil constituents as inputs. Three methods were used to predict volumetric moisture retention: (i) mid‐infrared (MIR) spectra and partial least squares regression, (ii) a neural network pedotransfer function (PTF) using laboratory particle‐size distribution and bulk density data, and (iii) a pedotransfer function with MIR‐predicted particle‐size distribution and bulk density as inputs. We used Lin's concordance correlation coefficient as a goodness‐of‐fit measure. Predictions of volumetric moisture retention on intact structured soils were generally poor, particularly at the wet end. Improved predictions were observed at dry‐end matric potentials, where moisture retention was more correlated with particle‐size distribution than soil structure. The neural network PTF was found to have better goodness of fit for all matric potentials; however, predictions at larger matric potentials (wet end) were still poor due to poor bulk density predictions. In light of this work, we propose that while MIR spectroscopy may be a valuable predictor of fundamental soil constituents such as particle‐size fractions and organic C, predictions of soil properties dependant on soil structure, such as volumetric moisture retention, may prove difficult. Mid‐infrared spectroscopy in combination with PTFs should provide improvements to moisture retention predictions through improved representation of the influential processes, namely soil structure and adsorptive forces.

Prediction of Yellow-Poplar (Liriodendron Tulipifera) Veneer Stiffness and Bulk Density Using near Infrared Spectroscopy and Multivariate Calibration

This study investigated the feasibility of using near infrared (NIR) spectroscopy and multivariate calibration to predict bulk density and stiffness of 3.2 mm thick yellow poplar veneer strips. Full-range (800–2500 nm) raw NIR spectra and spectra pre-treated using the first derivative method, along with spectra from three other different wavelength windows of 1200–2400 nm, 1800–2400 nm and 1400–2000 nm were regressed against the bulk density (kg m−3) values and the dynamic modulus of elasticity (stiffness; GPa) of the veneers using the projection to latent structures (PLS) method to develop calibration models. All predictive models developed performed well in the prediction of bulk density and stiffness of new test samples that were not included in the calibration models. R2 values ranged from 0.67-0.78 and 0.56-0.72, respectively, for bulk density and stiffness. There was significant improvement in models developed with first derivative spectra over models developed with raw spectra. The models developed using the first derivative used fewer latent variables to achieve predictive models with higher R2 values, lower root mean square errors of prediction (RMSEP) and standard errors of prediction (SEP). Models developed using the full NIR spectral range (800–2500 nm) and the NIR spectral region of 1200–2400 nm performed better than models developed using the restricted NIR wavelength regions of 1800–2400 nm and 1400–2000 nm. However, there was no clear distinction between models developed using the full NIR spectral range and the NIR spectral region of 1200–2400 nm. Overall, models developed with the first derivative pre-processed spectra using the whole NIR spectrum performed best in predictability. The results of this study show the potential of using multivariate data analysis coupled with NIR spectroscopy for on-line sorting and assessment of veneer stiffness prior to the lay-up process in the manufacturing of veneer-based engineered wood products such as plywood, Parallam and laminated veneer lumber.

Modeling and simulation of materials synthesis: Chemical vapor deposition and infiltration of pyrolytic carbon

Chemical vapor deposition and infiltration of pyrolytic carbon for the production of carbon fiber reinforced carbon is studied by modeling approaches and computational tools developed recently. Firstly, the development of a gas-phase reaction mechanism of chemical vapor deposition (CVD) of carbon from unsaturated light hydrocarbons (C 2H 4, C 2H 2, and C 3H 6) is presented. The mechanism consisting of 827 reactions among 227 species is based on existing information on elementary reactions and evaluated by comparison of numerically predicted and experimentally determined product compositions taking into account 44 stable gas-phase compounds formed in a tubular flow reactor. Secondly, a model and a computer code for two-dimensional transient simulations of chemical vapor infiltration (CVI) from methane into carbon fiber reinforced carbon are presented. The chemical model is based on a reduced multi-step reaction scheme for pyrolytic carbon deposition, which is derived from a mechanism based on elementary reactions, and a hydrogen inhibition model of pyrolytic carbon deposition. The coupled governing equations of mass transfer, chemical vapor deposition, surface growth, and gas-phase and surface chemical reactions are numerically solved by a finite element method (FEM). The computer code is applied to reveal densification processes of felts with fiber volume fractions of 7.1% and 14.2%. Numerically predicted bulk density distributions agree well with experimental results.

An application of genetic algorithms to the robust estimation of soil organic and mineral fraction densities

The development and use of a program that combines genetic algorithms and the least median of squares (LMS) method is reported. The combination of these two methods should provide a very robust approach for optimising non-linear functions. The program was specifically developed to estimate soil bulk density using a mass fraction model and was tested against a simulated dataset before being applied to the prediction of bulk density from a dataset created by combining measurements from a number of studies. The prediction of bulk density has a very important role in the calculation of soil elemental pools, and the predictive ability provided by the models described can be used to improve estimates of the key parameters such as soil carbon storage.

Thermal Conductivity of Some Compacted Trinidadian Soils as affected by Peat Content

The effects of incorporating peat into three Trinidadian agricultural soils before compaction on soil thermal conductivity k was investigated in a laboratory experiment. A factorial experiment was used to study the effect of peat applied at four levels (0%, 4%, 8% and 12% by mass) on maximum thermal conductivity k max and the corresponding moisture content on three soils (sandy loam, clay loam and clay) compacted using 5, 15 and 25 Proctor hammer blows. The compaction tests were carried out at different moisture contents ranging from 5% to 55%. Bulk densities achieved during soil compaction were measured to assist in the interpretation of results. Results showed that while mean values of k max of the soils declined significantly (probabilty P =0·001) from 1·49 to 1·04 W m −2 °C −1 with increasing peat content, the mean values of the moisture contents at which they occurred increased from 26·1% to 41·7%. While k max increased, the corresponding moisture contents decreased with increasing compaction levels. This applied to all soils tested. The clay soil had significantly lower values of thermal conductivities and greater corresponding moisture contents than the clay loam and the sandy loam soils. During soil compaction, thermal conductivity and bulk density behaved in the same manner, except that the maximum bulk density was achieved at lower moisture content than thermal conductivity. There was a significant interaction effect between soil type and peat content on thermal conductivity. There was also a reasonable agreement between the laboratory measured thermal conductivity data and those predicted using the Campbell model. A linear equation developed to relate thermal conductivity to bulk density suggests that a unique relationship may exist between the two parameters, when compacted soils are involved. A multiple linear regression equation was further developed for predicting bulk density from experimental factors.

Revisiting the hexane-water interface via molecular dynamics simulations using nonadditive alkane-water potentials

We present results addressing properties of a polarizable force field for hexane based on the fluctuating charge (FQ) formalism and developed in conjunction with the Chemistry at Harvard Molecular Mechanics (CHARMM) potential function. Properties of bulk neat hexane, its liquid-vapor interface, and its interface with a polarizable water model (TIP4P-FQ) are discussed. The FQ model is compared to a recently modified alkane model, C27r, also based on the CHARMM potential energy function. With respect to bulk properties, both models predict bulk density within 1%; the FQ model predicts the liquid vaporization enthalpy within 2%, while the C27r force field underestimates the property by roughly 20% (and in this sense reflects the quality of the C27r force field across the spectrum of linear and branched alkanes). The FQ hexane model realistically captures the dielectric properties of the bulk in terms of a dielectric constant of 1.94, in excellent agreement with experimental values in the range of 1.9-2.02. This behavior is also in conformity with a recent polarizable alkane model based on Drude oscillators. Furthermore, the bulk dielectric is essentially captured in the infinite frequency, or optical, dielectric contribution. The FQ model is in this respect a more realistic force field for modeling lipid bilayer interiors for which most current state-of-the-art force fields do not accurately capture the dielectric environment. The molecular polarizability of the FQ model is 11.79 A3, in good agreement with the range of experimental and ab initio values. In contrast to FQ models of polar solvents such as alcohols and water, there was no need to scale gas-phase polarizabilities in order to avoid polarization catastrophes in the pure bulk. In terms of the liquid-vapor and liquid-liquid interfaces, the FQ model displays a rich orientational structure of alkane and water in the respective interfacial systems, in general conforming with earlier simulation studies of such interfaces. The FQ force field shows a marked deviation in the interfacial dipole potentials computed from the charge densities averaged over simulation trajectories. At the liquid-vapor interface, the FQ model predicts a potential drop of -178.71 mV in contrast to the C27r estimate of -433.80 mV. For the hexane-water interface, the FQ force field predicts a dipole potential drop of -379.40 mV in contrast to the C27r value of -105.42 mV. Although the surface dipole potential predicted by the FQ model is roughly 3.5 times that predicted by the C27r potential, it is consistent with reported experimental potentials across solvated lipid bilayers in the range of 400-600 mV.

Using Soil Physical and Chemical Properties to Estimate Bulk Density

A stepwise multiple regression procedure was developed to predict oven-dried bulk density from soil properties using the 1997 USDA-NRCS National Soil Survey Characterization Data. The database includes both subsoil and topsoil samples. An overall regression equation for predicting oven-dried bulk density from soil properties (R2 = 0.45, P < 0.001) was developed using almost 47000 soil samples. Partitioning the database by soil suborders improved regression relationships (R2 = 0.62, P < 0.001). Of the soil properties considered, the stepwise multiple regression analysis indicated that organic C content was the strongest contributor to bulk density prediction. Other significant variables included clay content, water content and to a lesser extent, silt content and depth. In general, the accuracy of regression equations was better for suborders containing more organic C (most Inceptisols, Spodosols, Ultisols, and Mollisols). Bulk density was poorly predicted for suborders of the Aridisol and Vertisol orders which contain little or no organic C. Although organic C was an important variable in the suborder analysis, water content explained most (>30%) of the variation in bulk density for Udox, Xererts, Ustands, Aquands, and Saprists. Relationships between bulk density with soil volume measured on oven-dried natural clods and bulk density with soil volume measured at field-moisture content and one-third bar were also determined (R2 = 0.70 and 0.69, respectively; P < 0.001). Utilizing the regression equations developed in this study, oven-dried bulk density predictions were obtained for 71% of the 85608 samples in the database without bulk density measurements. While improving on methods of previous analyses, this study illustrates that regression equations are a feasible alternative for bulk-density estimation.

EFFECT OF MOISTURE CONTENT AND BROKEN KERNELS ON THE BULK DENSITY AND PACKING OF CORN

Shelled yellow dent corn samples were conditioned to three moisture content levels (12%, 15%, and 18% w.b.)and mixed with a prescribed amount of broken corn particles of known size (geometric mean diameter of 1.0, 1.4, 2.0, 2.8,and 4.0 mm) and concentration (2.5%, 5.0%, and 7.5% by weight) levels. The initial bulk density and grain compaction undersimulated overburden pressure tests were determined for each sample. Uniaxial compression tests were performed for sevenvertical pressure levels (3.4, 6.9, 14, 28, 55, 110, and 165 kPa) with a minimum of three replications each. Tests wereperformed at two locations with identical apparatus, which was fully described by Thompson and Ross (1983). These devicesused compressed air injected beneath a rubber diaphragm to apply vertical pressure uniaxially to a volume of granularmaterial. Deflections of the grain mass were measured with a dial gauge and were used to calculate changes in bulk densityand grain packing. Statistical models were tested for the initial bulk density and packing factor as a function of moisturecontent, broken corn particle size, and broken corn concentration level and their interactions. For clean corn, the initial bulkdensity was inversely affected by grain moisture while packing increased slightly with grain moisture. For corn mixed withfines, the initial bulk density decreased with grain moisture and the interaction of broken corn particle size and concentrationbut increased with the interaction of grain moisture and concentration of fines. Packing of corn mixed with fines increasedslightly with grain moisture and broken corn concentration. For a given pressure, the predicted bulk density from thedeveloped model was within 4% of the observed value, which was within the variation among test replications and may infact represent observed differences in bulk density caused by bin loading methods that have been reported by other engineers.The results can improve predictions by WPACKING, the ASAE standard for estimating capacities of cylindrical grain storagestructures.

CALIBRATION METHODS FOR NONDESTRUCTIVE MICROWAVE SENSING OF MOISTURE CONTENT AND BULK DENSITY OF GRANULAR MATERIALS

The principles of a nondestructive multiparameter microwave sensor operating at a single frequency are presented.Two calibration methods are reviewed and compared for simultaneous, independent determination of bulk density andmoisture content in granular materials. The first calibration method is based on measurement of attenuation and phase shiftthrough a layer of material in free space. The attenuation and phase shift were used to determine simultaneously bulk densityand moisture content, and the attenuation-to-phase ratio was used to determine moisture content independent of bulk densitychanges. For the second calibration method, relationships are established between the dielectric properties and bulk densityand moisture content. The dielectric properties were used to determine simultaneously bulk density and moisture content, anda calibration function, expressed in terms of these properties, was used to determine moisture content independent of bulkdensity. In addition, a complex-plane representation of the dielectric properties provided bulk density independent of temperatureand moisture content. Performance of these calibration methods in predicting bulk density and moisture content wascompared for wheat, corn, and soybeans at 10.0 GHz at room temperature. Bulk density and moisture content calibrationequations are given for each material along with corresponding standard errors of calibration.

Predicting bulk density of Ohio Soils from Morphology, Genetic Principles, and Laboratory Characterization Data

A 937‐horizon data set composed of site characteristics, morphology, and laboratory characterization data for soils of Ohio was used to develop soil bulk density (Db) prediction models. We tested the hypothesis that using a combination of continuous variables (laboratory data) and nominal variables (site/state factor and morphological class descriptors) would enable the development of improved Pedo‐Transfer Functions (PTFs) for Db Three primary models were developed. The Lab Model, composed entirely of continuous variables, accounted for 56% of the variability in Db Using only state factors and morphology as nominal variables, the Field Model explained 69%. A combined Field + Lab Model accounted for 72%. Restricting the data set to samples derived from loess and glacial till generated a Field + Lab Model that explained nearly 80% of the variability in Db for a subset of 402 horizons.

A comparison of soil maps, kriging and a combined method for spatially predicting bulk density and field capacity of ferralsols in the Havana–Matanzas Plain

Predictions of soil bulk density and field capacity at 0–10, 10–20 and 20–30 cm depths in Rhodic and Xhantic Ferralsols at the Havana–Matanzas plain are validated at a number of measured sites. It was done in order to compare the prediction accuracy under the constraint of a small data set by the following three procedures: (1) soil maps, (2) kriging, considering Xhantic and Rhodic Ferralsols as different soil types, as well as Ferralsols as a combined single soil type and (3) combined kriging–soil map predictions. The results show a considerable bias in the predictions made with the soil map, which is not found in kriging predictions. Generally, soil map predictions are also less accurate. However, the use of soil maps in Xhantic Ferralsols field-capacity predictions was the most reliable approach. The use of semivariograms for each soil type in kriging predictions only yields more accurate results for Rhodic Ferralsols, where enough data is available. The combined kriging–soil map procedure yields the smallest bias. Predictions of the combined procedure are more accurate, although accuracy differences found with the other two approaches were not very large. The combined kriging–soil map approach yields better predictions than the others for this case study.