Vol. 9(12), pp. 1025-1035, 27 March, 2014 DOI: 10.5897/AJAR2013. 8316 Article Number: 89CBF7246987 African Journal of Agricultural ISSN 1991-637X Copyright ©2014 Research Author(s) retain the copyright of this article http://www.academicjournals.org/AJAR

Full Length Research Paper

Spatial variability of color parameters and soil properties in an alluvial soil

Fevzi AKBAS

Soil, Water and Desertification Research Institute Konya, Turkey.

Received 2 December, 2013: Accepted 21 March, 2014

Understanding of spatial variability of soil properties provides the factors and processes controlling their potential in agricultural production. The objectives of this study were to analyze spatial structure of soil properties and soil color parameters (L*, a*, b*) in an alluvial field (45 ha) under different crop patterns, and to map soil properties using geostatistical methods. A geostatistical sampling scheme was adopted, and 188 soil samples were taken from (0-20 cm) and (20-40 cm). Soil color value b* showed the lowest variation (CV = 6.4% for topsoil and CV = 6.1% for subsoil) between soil color parameters. Available Phosphorus showed the highest variation (CV = 64.6% for topsoil and CV = 52.4% for subsoil). Geostatistical range varied from 114 m (available phosphorus) to 633 m (EC) in topsoil. Soil pH had the shortest (72 m), and CaCO3 had the longest range value (612 m) in subsoil. Map of CaCO3 content visually correlation with lightness (L*) map in both depths but maps of the organic matter did not show visual correlation with maps of L* in both depths. Organic matter maps can be used for site specific N application because OM to some extent characterizing potential N-supplying capacity could be used as reference for N fertilizer recommendation.

Key words: Geostatistics, semivariogram, kriging, soil color.

INTRODUCTION

Different parent materials and topographies cause spatial of soil (Sanchez-Maranon et al., 2004; Bhavanarayana, variations, but variation in soil properties is also a result 2001). Because color is easily perceived in the field and of a number of long-term and short term soil is related to other soil properties, it has often been used management strategies. Understanding the distribution of as a rapid assessment indicator of such thing as drainage soil properties in the field scale is essential in refining class, , and soil organic carbon which is agricultural management practices (McBratney and affected to usage of variable rate fertilizer, pesticides Pringle, 1999) while minimizing environmental damage. application (Konen et al., 2003; Barrett, 2002). Color is one of the most apparent and most significant Properties such as soil organic carbon content, iron characteristics of soil, consistently used as a definitive content, content, and texture have been shown criterion in the identification, classification and separation to have a good correlation with soil color. with dark

*Corresponding author. E-mail: [email protected] Author(s) agree that this article remain permanently open access under the terms of the Creative Commons Attribution License 4.0 International License 1026 Afr. J. Agric. Res.

surface horizons are generally linked with high organic Fleming et al. (2004) and Khosla et al. (2002) were matter content, classifying them as fertile and suitable for delineated from the variability in color observed in bare plant growth. Dark brown and black soils are also thought soil imagery of conventionally tilled field. The color of to contain high level of nitrogen, have good aeration and bare soil surface is largely governed by soil organic drainage, and pose low erosion risk. Generally, the matter and moisture. Fleming at al. (2004) used soil color opposite is thought of light colored soil. The color of iron with farmer experience one of the methods delineating containing soil that undergo oxidation and management zones in two center-pivot irrigated fields. reduction reactions can provide useful information on the They found that this methods promising in the future hydrologic condition of soil (Viscarra et al., 2006). application. Most studies on soil color employed the Munsell color In agricultural land, to reduce the variability in yield and system because of its ease of use, ready availability, and to improve crop quality, variable rate technology needed. historical importance to . However, this In addition to improvement of crop yield efficiency, method is semi quantitative, because it allows a economical issues and environmental concerns has subjective match of soil samples to the Munsell color necessitated the introduction of variable rate technology. chips (Konen et al., 2003; Barrett, 2002). Newer color Variable rate technology (site specific recommendations) systems, like those created by the Commission is demanded for the determination of spatial variability Internationale d’Enclairage (CIE) may be better suited to precisely on farm level. The objectives of this study were represent a uniform color space, that is, a color space to: (i) quantify the spatial variability and the pattern of soil where the Euclidean differences between colors are properties including soil color parameters and (ii) equivalent to the human perceptions of those differences generate maps for the spatial distribution of these (Melville and Atkinson, 1985; Scheinost and variables to manage agricultural inputs. Schwertmann, 1999). Barrett (2002) used spectrophotometer to quantify soil color in situ and MATERIALS AND METHODS reported that spectrophotometer provided a higher degree of precision. Some researcher used soil CIELAB Description of the study area color parameter to represent soil color and reported relationships between soil color parameter and soil The study site is recently consolidated area and located in 56 km properties (Sánchez-Marañón, 1997; Konen et al., 2003; north east of the Tokat province within the middle Black Sea region of Turkey (40° 28' 41"N, 36° 59' 25" E). Study area is located Gunal et al., 2008). between Kelkit River and Canik Mountains. The field is relatively Spatial distribution of soil properties is not uniform. This flat; having slope of 1 to 2% and a small creek called Fatli encircles irregular distribution of soil characteristics, such as the eastern border of the study area. Coarse particles and gravels nutrient availability, organic content, and content, found at 50 cm of the soil profiles in the middle of the study area, implicitly reflects the processes that occur within the indicating that this small creek probably changed its bed several larger ecosystem (Corstanje et al., 2006). The times in the past. The soils studied were classified as Vertic Hapluestepts ( Staff, 1999) based on a cambic B geostatistical approach treats soil properties as horizon and vertic properties observed during sampling. continuous variables and models, these as the most likely The mean annual precipitation and air temperature are 456.4 mm outcomes of random processes (Webster and Oliver, and 12.3°C, respectively. Land leveling and consolidation 2007), which allows random variation in soil properties to operations were completed in the study area during 1999 and 2000 be formulated mathematically. Besides to this basic growing seasons. A small ridge in the middle of the field was also leveled during this operation. The rotation of wheat, sugar beet, definition, Geostatistics is practically concerned with maize (second crop), and alfalfa has been applied for a long time in detecting, estimating and mapping the spatial patterns of the field. At the sampling time, some portions of the study area had regional variables, and uses point information for been planted with winter wheat, sugar beet, and alfalfa, and rest of interpolation. Semivariogram is the core of geostatistical the field was prepared for tomato production (Figure 1). In wheat studies and this instrument distinguishes variation in and sugar beet production, diammonium phosphate (DAP) and measurements separated by given distances (Goovaerts, ammonium nitrate (AN) as a base fertilizer, and urea as a top fertilizer were used. DAP and urea fertilizers were also commonly 1997; Isaaks and Srivastava, 1989; Rossi et al., 1992). used for tomato production as a source of P and N in the study Semivariogram models provide the necessary information area. for kriging, which provides the best linear unbiased prediction at un-sampled locations. In practice, kriging is often a precursor to management decisions (Lark and Sampling design and analysis

Ferguson, 2004). Geostatistics have been used to The sampling area is about 54 ha, and boundary of area is given in estimate soil physical (Jung et al., 2006; Iqbal et al., Figure 1. Systematic grid sampling was applied in 100 to 100 m grid 2005), biological, and ecological properties (Rochette et cells. Soil samples were collected from upper nodes of each grid in al., 1991; Rossi et al., 1992; Gaston et al., 2001). Some May, 2005. At the grid intersection points, 44 main sampling points work has optimized sampling strategies (Buscaglia and were created in 100 to 100 m grid design. In order to model short range variability of soil attributes, ten fine- transects with 2, 5, 10, Varco, 2003; Ruffo et al., 2005). 25 and 50 m intervals were sampled. The fine transects were Site-specific management zones as described by located regularly in the north-south and east-west directions, Akbas 1027

Figure 1. Layout of study area and sampling design.

superimposing them on the lines between the grid nodes (Figure 1). accurate, and precise than color quantification by the human eye Soil samples were collected at two depths ( topsoil, 0-20 cm and using Munsell soil color books or color chart developed specifically subsoil, 20-40 cm) at 94 sampling sites, and the total number of soil for estimating soil organic carbon or soil OM concentrations (Konen samples was 188 (Figure 1). et al., 2003). Colors are specified in CIELAB color space using The texture of <2 mm fraction of each sample (% sand, % and three coordinates: L* (similar to munsell value), a* (denoting hue on % ) was determined by the Bouyoucos hydrometer method the redgreen axis), and b* (which denotes hue on the yellowblue (Gee and Bouder, 1986). Electrical conductivity and pH of soils were axis). The L* axis represent lightness ranging from no reflection for measured 1:2.5 soil water suspension (Hendershot et al., 1993). black (L=0) to perfect diffuse reflection for white (L=100). The a* Organic matter was measured by the Walkey Black method (Nelson axis is the redness ranging from negative values for green to and Sommers, 1982), and calcium carbonate content was positive values for red, and b* axis is the yellowness ranging from determined with a pressure calcimeter (Nelson, 1982). Total N was negative values for blue and positive values for yellow (Melville and measured Kjeldal method (Bremner, 1965), and available Atkinson, 1985). phosphorus was determined with Olsen et al. (1954). The colors of soil samples were measured with the CR300 Soil color is commonly and widely measured using Munsell soil Chroma Meter (Minolta, Osaka, Japan) and spectral reflectance color chart (Munsell Color Company, 1994) Because of variation in was determined over the 400 to 700 nm range. We calibrated the light sources and surface properties soil color chips and soil, chroma meter with a standard white plate (Minolta) at the start of observers perceive varied and assignment of soil color. The each sample set run. Soils were ground and passed through a 2 limitations can be eliminated by using a colorimeter, which is an mm sieve, and placed in a Petri dish to provide minimum of a 1 cm instrument capable of providing both accurate and precise soil color thickness, and five spectra from different locations were averaged data (Torrent and Barron, 1993). from each sample. Colorimeters have numerous advantages for quantifying soil color properties. The technique is non-destructive and further analyses can be run on the same sample. The instrument is Data analysis portable and can be taken to the field. Color parameter may be determined on both dry and moist samples in seconds. Sample Descriptive statistics was conducted on physical and chemical preparation is minimal and the methods is more consistent, properties of soils studied. Exploratory data analysis was made 1028 Afr. J. Agric. Res.

calculating minimum, maximum, arithmetic mean, standard where z(xi) is the observed value at location i, ẑ(xi) is the predicted deviation, coefficient of variation (CV), skewness, and kurtosis for value at location i, and n the sample size. Small MAE values each variable. Classical statistic was conducted with SPSS 10.0 indicate a method with errors, overall. MSE, which is a measure of statistical software (SPSS, 2000). The spatial structure of variables the sum of the squared residuals, was characterized using experimental semivariogram, expressed as; n 1 2

MSE  z(xi )  zˆ(xi ) 2 n 1 N h i1 (5) (1) h  zxi  z xi  h  2N(h) i1 is the difference at any point which gives an indication of the magnitude of differences such that small MSE values indicate more

accurate prediction, point-by-point. Another measure of where, h is the experimental semivariogram value at the effectiveness of prediction is the G value (goodness of prediction), distance interval h; N(h) is number of sample value pairs within the n n distance interval h; z(xi) and z (xi+h) are the sample values at two    G 1 z(x ) zˆ(x ) 2 / z(x ) z 2  *100 points by the distance interval h (Isaaks and Srivastava, 1989).    i  i i   Nugget effect expressed as (Co/Co+C), where Co is nugget   i11i  (6) variance and Co+C is sill, quantifies spatial dependency of the soil properties (Cambardella et al., 1994). Anisotropic semivariograms where z is the sample mean. When the G value equal the 100% did not show any differences in spatial dependence based on indicates perfect prediction, negative values indicates the directions, hence isotropic semivariograms were considered in predictions are less reliable (Schloeder et al., 2001). further analyses. The parameters of model: nugget semivariance, range, and sill were determined. Nugget semivariance is the variance at zero distance; sill is the semivariance value at which the RESULTS AND DISCUSSION variogram levels off; and range is the distance at which one variable become spatially independent or the lag distance at which Descriptive statistics the semivariogram reaches the sill value. Modeling of isotropic + experimental semivariogram was performed with GS (Version 7) The descriptive statistics for soil parameters in the study statistical software (Gamma Design Software, 2004), and a maximum lag distance of 700 m were applied in experimental area are presented in Table 1. The means of soil semivariogram modeling. Since the use of log-transformed data variables were similar in two depths suggested that slightly affected the kriging predictions, untransformed data were surface soils (0-20 cm) of the study area generally look used in predictions (except available phosphorus in subsoil). The like deeper part (20-40 cm). The soils were clayey in best fit model selection for experimental semivariogram was done texture, the mean clay content 42.87 and 42.90% in by the leaving-one-out method of cross validation. We created interpolation maps of each variable through ordinary kriging method topsoil and in subsoil, respectively. Average calcium using their respective isotropic semivariogram models (Isaaks and carbonate content of the topsoil was 8.77%, which was Srivastava, 1989). slightly lower than that of subsoil (9.20%). Organic matter Ordinary kriging method assumes that the data set has a (OM) content of soils investigated was low due to the stationary variance but also a non-stationary mean within the intensive tillage by moldboard, and absence of manure search radius. Ordinary kriging is by far the most common type of and plant residue additions. OM content of topsoil kriging in practice and recommended for most data set. We (1.79%) was higher than that of subsoil (1.25%). The pH estimate Z at a point x0 by Ẑ (x0), by was similar in top and subsoil, 7.96 and 8.08, respectively  N (Ozgoz et al., 2009), and soluble salt content was low at Z ( x 0 )    i z ( x i ) both depths (Table 1). Total N (TN) of subsoil is slightly i 1 (2) lower than topsoil. Available phosphorus (AP) is almost 3 times higher in topsoil than in subsoil. This indicates that A weighted average of the data, where i are the weights. The excessive phosphorus fertilization is subject to surface ensure that the estimate is unbiased, the weights are made to sum to 1. soil of study area. The color of samples was between 3.99 and 5.56 for a*, between 12.75 and 16.75 for b*, and N between 31.31 and 45.73 for L* in topsoil. Mean color i  1 values of a*, b*, and L* were slightly higher in subsoil. i1 (3) Majority of the soil properties had positive skewness and kurtosis values in both depths. All variable except , and the expected error is zero. At least 12 neighboring points were EC and AP in topsoil were slightly skewed (skewness < considered for ordinary kriging interpolations and the maps of soil 1). In subsoil sand, CaCO3, and AP had skewness value properties were generated with Geostatistical Extension of ArcGIS 8.1 higher than 1 and other soil variables were slightly (ESRI, 2001). The mean absolute error (MAE) and the mean squared error skewed. (MSE) were calculated to compare prediction. The MAE is a measure of Soil pH had the lowest CV (1.3%) in topsoil and the sum of residual, subsoil. While silt, L*, a* and b* had low CV values (6.4 to

12.1%), clay, sand, CaCO , OM, EC and TN had medium 1 n 3 MAE  z( x )  zˆ( x ) CV values in topsoil (21.8 to 32.6%). AP had the highest  i i -1 n i 1 (4) CV (64.6%) and ranged from 5.26 to 105.25 mg kg in Akbas 1029

Table 1. Descriptive statistics of soil properties of topsoil and subsoil.

Mean Std. Dev. Minimum Maximum Skewness Kurtosis CV (%) Property 0-20 20-40 0-20 20-40 0-20 20-40 0-20 20-40 0-20 20-40 0-20 20-40 0-20 20-40 cm cm cm cm cm cm cm cm cm cm cm cm cm cm Clay (%) 42.87 42.90 9.35 10.85 17.24 6.93 56.41 62.23 -0.72 -0.81 -0.45 0.28 21.8 25.2 Silt (%) 30.70 29.50 3.72 4.46 23.42 16.49 43.30 41.24 0.43 -0.04 0.36 0.38 12.1 15.1 Sand (%) 26.43 27.60 8.62 10.49 14.72 14.72 55.96 76.58 1.13 1.68 0.57 4.03 32.6 38.0

CaCO3 (%) 8.77 9.20 2.45 3.07 4.74 4.37 15.50 18.41 0.58 1.04 -0.46 0.73 27.9 33.3 Org. Mat. (%) 1.79 1.25 0.49 0.45 0.58 0.19 3.22 2.36 0.51 -0.12 0.59 -0.13 27.3 36.0 pH 7.96 8.08 0.11 0.11 7.63 7.82 8.24 8.51 0.05 0.26 0.59 1.80 1.3 1.3 EC dSm-1 0.57 0.44 0.18 0.07 0.28 0.22 1.15 0.64 1.14 0.05 1.02 0.77 31.3 16.9 AP mgkg-1 32.94 11.73 21.29 6.15 5.26 4.85 105.25 32.40 1.19 1.11 1.15 0.66 64.6 52.4 TN (%) 0.09 0.07 0.02 0.02 0.03 0.02 0.13 0.16 -0.05 0.99 -0.19 1.83 25.0 28.5 L* 39.40 41.47 2.73 3.27 31.31 32.22 45.73 49.32 -0.42 -0.18 0.53 0.51 6.9 7.8 a* 4.74 5.08 0.33 0.40 3.99 3.77 5.56 6.39 0.05 0.26 -0.40 1.32 6.9 7.8 b* 14.77 15.35 0.95 0.94 12.75 13.09 16.75 18.10 -0.09 0.55 -0.50 0.30 6.4 6.1

Table 2. Correlations (r) between soil characteristics and color parameters (L*, a*, b*) in 0-20 cm and 20-40 cm soil depths.

Parameter CaCO3 Org. Mat. EC Clay Sand L* 0-20 cm 0.46** 0.10 -0.24* -0.12 0.15 20-40 cm 0.58** -0.19 -0.36** -0.16 0.00

a* 0-20 cm 0.46** 0.11 -0.28** 0.02 0.17 20-40 cm 0.61** -0.22* -0.18 0.26* 0.00

b* 0-20 cm 0.35** -0.05 -0.47** -0.42** 0.46** 20-40 cm 0.53** -0.31** -0.48** -0.34** 0.26*

*, ** Significant at 0.05 and 0.01 level of significance.

surface soil (Table 1). Sand and OM shifted from medium (P<0.01). Soil color parameter L* and a* were not to high variability in subsoil compared to topsoil based on significantly correlated with clay and sand. But clay and CV values which were 38 and 36%, respectively. Other b* negatively correlated in topsoil and subsoil (r= -0.42** soil properties were in the same variability class in topsoil and r= -0.34**) and sand had positive correlation with b* and subsoil. Management related soil properties EC and in topsoil (P<0.01). AP had higher CV in topsoil while OM and total N had slightly higher variability (higher CV) in subsoil than topsoil. Table 2 shows the degree of correlations Geostatistical analysis between the selected soil properties. Strong correlation was found between soil color parameters, L*, a* and b* Table 3 lists the best-fitted isotropic semivariogram model with calcium carbonate in topsoil and subsoil (P<0.01). parameters and some spatial structural indices of soil Correlation coefficient between CaCO3 and soil color properties in topsoil and subsoil. The directional parameter (L*,a*,b*) were higher in subsoil compared to semivariograms calculated at the angles of 0° (N - S), 45° topsoil. Soil color b* values inversely correlated with OM (NE - SW), 90° (E - W), and 135° (SE - NW) for the (r = -0.31**, P<0.01) in subsoil. Correlation coefficient measured variables indicated no distinct anisotropy. between EC with L* was negative in subsoil and there Therefore, omni-directional semivariograms were was negative correlation between EC and a* in topsoil. In modeled to determine the spatially dependent variance both soil depth EC was inversely correlated with b* within the research area. Clay, sand, CaCO3 and b* were 1030 Afr. J. Agric. Res.

Table 3. Geostatistical parameters of soil properties in 0-20 cm and 20-40 cm soil depths.

-1 1 -1 Parameter Clay % Silt % Sand % CaCO3 % Org. Mat. % pH EC dSm AP mgkg TN (%) L* a* b* Surface soil (0-20 cm) Model Sph Exp Sph Sph Exp Exp Exp Exp Exp Exp Exp Sph

(Co) 12.5 6.92 3.5 1.5 0.1019 0.00575 18510 0.217 0.00019 2.49 0.0443 0.227

(Co+C) 102.6 14.37 88.8 7.18 0.2598 0.0116 39570 0.435 0.00038 6.557 0.1196 0.959 Range(m) 379 297 370 486 252 358 633 114 135 219 501 259

(Co)/ (Co+C) 12.1 48.1 3.9 20.9 39.2 49.6 46.8 49.8 50 37.9 37 23.7 r2 0.84 0.80 0.80 0.84 0.79 0.77 0.68 0.45 0.54 0.80 0.85 0.81 RSS3 1606 14 1991 8.68 0.006 8.5x10-6 2.2x108 0.03 1.2x10-8 3.56 0.001 0.16 SC2 S M S S M M M M M M M S MAE 3.74 2.76 2.96 1.04 0.33 0.08 0.13 15.89 0.01 2.01 0.21 0.60 MSE 27.11 11.83 20.36 2.02 0.20 0.01 0.03 453.8 0.0003 6.54 0.07 0.66 G 69 13 72 66 15 18 14 -1 6 11 58 26 Subsurface soil (20-40 cm) Model Sph Exp Sph Exp Exp Exp Exp Exp Exp Exp Sph Sph

(Co) 32.8 7.29 22.8 0.68 0.0357 0.00245 2090 0.052 0.00028 2.67 0.0369 0.263

(Co+C) 120.1 21.15 124 12.3 0.207 0.0117 5561 0.242 0.00057 10.63 0.1725 0.845 Range(m) 402 516 321 612 129 72 213 75 162 183 515 106

(Co)/ (Co+C) 27.3 34.4 18.3 5.5 17.2 20.9 37.5 21.5 49.1 25.1 21.4 31.1 r2 0.82 0.89 0.68 0.95 0.69 0.50 0.70 0.87 0.41 0.73 0.82 0.86 RSS3 2567 26.4 5115 13.8 9.8x10-3 2.3x10-5 3.9x106 2.7x10-3 6.9x10-8 15.1 5.3x10-3 0.09 SC M M S S S S M S M M S M MAE4 5.51 2.96 4.64 1.20 0.30 0.08 0.05 4.48 0.02 2.18 0.229 0.2,69 MSE4 62.48 15.99 54.22 2.90 0.14 0.01 0.005 34.73 0.001 8.03 0.09 0.82 G4 46 19 50 69 28 6 13 7 -7 2 39 6

1log transformed data were used, 2SC: Spatial class; S: strong; M: moderate, 3 RSS: Residual sum of squares, 4MAE: Mean absolute error, MSE: Mean squared error; G: Goodness of prediction as definedby Schloeder at al. (2001).

modeled quite well with spherical model, whereas soil properties. A variable has strong spatial (Table 3). Clay, sand, CaCO3 and b* were the remaining properties were modeled with dependency if the ratio less than 25%, moderate strongly spatially dependent, with the nugget/sill exponential models in topsoil. Spherical model spatial dependency if the nugget/sill is between 25 ratio ranging from 3.9 to 23.7% in topsoil. Silt, were used for clay, sand, a* and b* in topsoil while and 75%, and weak spatial dependency for OM, pH, EC, AP, TN, L* and a* were moderately exponential model was provided the best fit for nugget/sill are greater than 75% (Cambardella et spatially dependent and nugget/sill values varied other soil properties in subsoil. al., 1994). The resulting semivariograms indicated between 37 and 49.6% in topsoil. While sand, The nugget/sill ratio (Co/C+Co) can be regarded the existence of moderate to strong spatial CaCO3, OM, pH and a* were strongly spatially as a criterion to classify the spatial dependency of dependence for all soil properties for each depth dependent, the remaining soil properties were Akbas 1031

moderately spatially dependent in subsoil. The spatial ranges of L* and b* decreased with depth, from 219 to variability of soil properties may be affected by both 183 m in the plow layer for L* and from 259 to 106 m at intrinsic (soil formation factors such as deeper depth for b*. But the range values of a* were very and climate) and extrinsic factors ( close to each other in both depths and these range practices, such as fertilization and irrigation). In general, values were 501 m in topsoil and 515 m in subsoil. strong spatial dependency of soil properties can be Fine textured soil was found in the north and north- attributed to intrinsic factors, and weak spatial central where large areas within the range of 45 to 55% dependency can be attributed to extrinsic factors in clay content were common. Reversely, sandy soil were (Cambardella et al., 1994). Textural components (clay located south part of the study area. Based on contour and sand) and CaCO3 showed strong dependency in maps patterns in Figure 2, EC, TN and AP were patchy surface soil. When soil properties show strong spatial spatial distribution in subsoil but more regular patterns dependency it may indicate that the variability in these and steady structures in topsoil. It is somewhat surprising properties is controlled by intrinsic variation. Nugget/sill that no patch distribution of EC, TN and AP was noted in ratios of management related soil properties were higher this study for surface soil. It is reported that patchy in topsoil than subsoil. This was suggesting that the distributions of soil maps was mainly controlled by extrinsic factors such as fertilization, plowing and other random factor (Cambardella et al., 1994). The level of management practices may be weakened their spatial EC, TN and AP in surface soil is generally altered by correlation in topsoil. These soil properties were OM, pH, management (fertilizer application, irrigation, plant EC, AP, TN which showed moderate spatial dependency uptake) therefore patchy distribution may indicate in topsoil. AP and TN showed highest nugget effect in management effects to these properties. Some other topsoil and this was evidence effect of fertilizer studies reported that weak or no spatial dependency for application to the soil heterogeneity. Sampling area was TN (Eltaib et al., 2002; Jung et al., 2006) and for AP a 45 ha-field and covered with four different crop patterns (Brouder et al., 2001; Han et al., 2005). Spatial in about 32 production parcels (Figure 1). Various crop distribution patterns of clay, silt, sand and CaCO3 were patterns at the time of sampling associated with different almost same in topsoil and subsoil. rate of nitrogen and phosphorus fertilizer usage The distribution maps of all variables are shown in weakened spatial correlation of AP and TN in topsoil. Figure 2. CaCO3 content visually correlated with soil Nugget variance (Co) represents variance due to lightness (L*). Significant correlation existed between measurement error or short range variability of the lightness (L*) and CaCO3 in both depth (Table 2). As property which cannot be detected with the current scale reported in previous studies (Sánchez-Marañón et al., of sampling. The semivariance increases as the 1997; Spielvogel et al., 2004) greater lightness values are separation distance between sample locations increases, expected for the soils with high CaCO3 content. High rising to an approximately constant value called sill calcium carbonate occurred in the northern area and low (Co+C). Spherical semivariogram model was used for the calcium carbonate in the southern. Northern part of the semivariograms of clay, sand, calcium carbonate and b* study area had calcium carbonate over 10% (up to 18.41) and exponential semivariogram model was used other and lightness values were also higher than other parts. soil properties in topsoil. Clay, sand and a* were best The corresponding L* values at this part of the study area modeled with spherical semivariogram model and were 45.7 in topsoil and 49.3 in subsoil. remaining properties were modeled with exponential Maps of the organic matter did not show visual model in subsoil. correlation with maps of L* at both depths. Organic One of the useful results that is obtained from spatial matter and soil lightness were reported inversely analysis is the range. This value is the approximate correlated (Viscarra Rossel et al., 2006; Sánchez- distance from one point to another within the field, which Marañón et al., 1997) but no significant correlation was would be assumed to be correlated. Therefore, a small also detected between OM and L* in this study (Table 2). value would indicate a great amount of variability within a Similar to L* (lightness) the a* values (red-green axis) the field. Large values indicate greater distances that the northern part of the study area higher a* values meant samples could be obtained and the data still be that soils of this region was slightly more redder. Coarse correlated. In our study, range values of all soil properties textured area (low in clay content high in sand content) in topsoil were greater than 200 m except AP and TN. In had higher b* values (yellow blue axis) it meant that the contrast to topsoil six soil properties (OM, pH, AP, TN, L* soils of this area were slightly more yellowish. However, and b*) had range values lower than 200 m in subsoil. the difference between maximum and minimum values of This indicates that the mentioned soil properties were soil color parameters a* and b* were very narrow more variable in subsoil compared to the topsoil. indicating soils were homogeneous in their redness and However, with the exception of pH and AP, the range of yellowness. This can be explained by the extent of the various semivariogram models which exceeded 100 m study area. The test area is relatively small (45 ha) for indicated the presence of spatial structure beyond the observing distinct parent material accordingly wide range original main grid sampling distance in the subsoil. The color differences in soils of the study area were not 1032 Afr. J. Agric. Res.

Akbas 1033

Figure 2. Spatial distribution maps of soil properties (L*, a*, b*, clay, silt, sand, CaCO3, OM, pH, EC, AP, TN). 1034 Afr. J. Agric. Res.

detected. For the first depth (0-20 cm) CaCO3 ranged Conflict of Interests from 4.74 to 15.50%, whereas in second depth (20-40 cm), CaCO3 ranged 4.37 to 18.41%, indicating similar The author(s) have not declared any conflict of interests. range of calcium carbonate content in the study area. According to correlation coefficients given in Table 2, CaCO3 content and soil color a* values positively REFERENCES correlated (P<0.01); in addition, these properties Barrett LR (2002). Spectrophotometric colour measurement in situ in exhibited largest and similar geostatistical range values well drained sandy soils. Geoderma 108:49- in both depth, indicating significant spatial structure 77.http://dx.doi.org/10.1016/S0016-7061(02)00121-0 extended over 500 m (Table 3). The spatial distribution of Bhavanarayana M (2001). Characterisation of soil colour and its use in CaCO and soil color a* value generally show a gradient estimation. Physical Methods of Soil Characterization 3 (Ed) J. Behari, Narosa Pub. House. New Delhi India, pp. 87-99. north to south, with CaCO3 elevated areas on the north Bremner JM (1965). Total Nitrogen. Editor: Black C.A. Methods of Soil edge of the study area (Figure 2). Gunal et al. (2008) Analysis. Part 2 American Society of Agronomy. Madison Wisconsin reported strong relationship between CaCO3 and soil USA, pp. 1149-1178. color a* value, and interpreted that the rubification Brouder S, Hofmann B, Reetz HF (2001). Evaluating spatial variability of soil parameters for input management. Better Crops 85:8-11. process that yielded the red color in soil was the result of Buscaglia HJ, Varco JJ (2003). Comparison of Sampling Designs in the the formation of iron oxides. Detection of Spatial Variability of Mississippi Delta. Soil Sci. Soc. Am. According to cross-validation results (Table 3), there J. 67:1180-1185.http://dx.doi.org/10.2136/sssaj2003.1180 were four prominent soil properties for which we obtained Cambardella CA, Moorman TB, Novak JM, Parkin TB, Karlen DL, Turco RF, Konopka AE (1994). Field-scale variability of soil properties in relatively more accurate and effective prediction based central Iowa soils. Soil Sci. Soc. Am. J. 58:1501-1511. on MAE, MSE and G values in both depths. These were http://dx.doi.org/10.2136/sssaj1994.03615995005800050033x clay, sand, CaCO3 and soil color a* value and calculated Corstanje R, Grunwald S, Reddy KR, Osborne TZ, Newman S (2006). G values of these properties were positive and a lot Assessment of the spatial distribution of soil properties in a Northern Everglades Marsh. J. Environ. Quality 35:938-949. higher. G values of other soil properties were positive, http://dx.doi.org/10.2134/jeq2005.0255 PMid:16641332 indicating interpolated maps represented a significant Eltaib SM, Soom MAM, Hanafi MM, Shariff ARM, Wayayok A (2002). improvement from the interpolation of the sample mean. Spatial variability of N, P, and K in rice field in Sawah Sempadan, G values AP in topsoil and TN in subsoil resulted in Malaysia. Songkhlanakharin J. Sci. Technol. 24:321-328. ESRI (2001). Arcgis 8.1 Environmental System Research Institute, negative values which indicates prediction maps are less Redland CA, USA. reliable. Schloeder et al. (2001) reported that poor Fleming KL, Heermann DF, Westfall DG (2004). Evaluating soil color interpolation results may be due to either very high with farmer input and apparent soil electrical conductivity for nugget or very high sill-to-nugget ratios (>30%). High management zone delineation. Agron. J. 96:1581-1587. http://dx.doi.org/10.2134/agronj2004.1581 nugget values yielded higher predictions than real values Gamma Design Software (2004). Geostatistics For The Environmental while high sill-to-nugget ratios resulted increase Sciences, Plainwell, –Michigan, USA. prediction variances. Both were consequences of limited, Gaston LA, Locke MA, Zablotowicz RM, Reddy KN (2001). Spatial high variable and weakly autocorrelated data (Schloeder variability of soil properties and weed populations in the Mississippi Delta. Soil Sci. Soc. Am. J. 65:449–459. et al., 2001). http://dx.doi.org/10.2136/sssaj2001.652449x Gee GW, Bouder JW (1986). Particle Size Analysis. In: A. Clute (ed.) Methods of Soil Analysis. Part I Agronomy No:9 American Society of Conclusion Agronomy. Madison, Wisconsin, USA. Goovaerts P (1997). Geostatistics for Natural Resources Evaluation. Oxford University Press New York0, P. 483. Coefficient of variation ranged from 1.3% for pH and to Gunal H, Ersahin S, Yetgin B, Kutlu T (2008). Use of chromameter- almost 64.6% for soil AP in topsoil indicating the measured color parameters in estimating color-related soil variables. heterogeneity of soil properties. However, soil color Communication Soil Sci. Plant Anal. 39:726- 740.http://dx.doi.org/10.1080/00103620701879422 parameters (L*, a* and b*) were much lower CV values Han XZ, Song CY, Wang SY, Tang CX (2005). Impact of long-term (6.4 to 6.9%), colors of study soils seemingly fertilization on phosphorus status in black soil. 15:319– homogeneous. Soil color parameters can be used to 326. delineate soil management zones. OM, TN and AP maps Hendershot WH, Lalande H, Duquette M (1993). Soil reaction and exchangeable acidity. ın soil sampling and methods of analysis. showed steady spatial pattern with regular borders was Carter M.R.(ed) Canadian Society of Soil Science, CRC Press Inc. very suitable variable rate fertilization to improve crop Boca Raton, Florida. USA. yield quality. OM and TN maps can be used for site Isaaks EH, Srivastava RM (1989). Applied Geostatistics, Oxford specific N application because OM and TN to some University Press, New York, USA, P. 561. Iqbal J, Thomasson JA, Jenkins JN, Owens PR, Whisler FD (2005). extent characterizing potential N-supplying capacity could Spatial variability analysis of soil physical properties of alluvial soils. be used as reference for N fertilizer recommendation. Soil Sci. Soc. Am. J. 69:1338- Similarly, AP maps in surface soil could be used as site 1350.http://dx.doi.org/10.2136/sssaj2004.0154 specific phosphorus management. Application variable Jung WK, Kitchen NR, Sudduth KA, Anderson SH (2006). Spatial characteristics of soil properties in an agricultural field. Soil rate fertilization reduces excessive usage of fertilizer and Sci. Soc. Am. J. 70:1387– so protect environment from oversupplied nutrients in 1397.http://dx.doi.org/10.2136/sssaj2005.0273 conventional farming systems. Konen ME, Burras CL, Sandor JA (2003). Organic carbon, texture, and Akbas 1035

quantitative color measurement relationships for cultivated soils in Monograph. 62:277–314.http://dx.doi.org/10.2307/2937096 North Central Iowa. Soil Sci. Soc. Am. J. 67:1823-1830. Ruffo ML, Bollero GA, Hoeft RG, Bullock DG (2005). Spatial variability http://dx.doi.org/10.2136/sssaj2003.1823 of the Illinois soil nitrogen test: implications for soil sampling. Agron. Khosla R, Fleming K, Delgado JA, Shaver T, Westfall DG (2002). Use J. 97:1485–1492.http://dx.doi.org/10.2134/agronj2004.0323 of site-specific management zones to improve nitrogen management Sánchez-Mara-ón M, Delgado G, Melgosa M, Hita E, Delgado R (1997). for precision agriculture. J. Soil Water Conserv. 57:513–518. Cielab color parameters and their relationship to soil characteristics in Lark RM, Ferguson RB (2004). Mapping the conditional probability of mediterranean red soils. Soil Sci. 162:8(11):833-842. deficiency or excess of soil phosphorous, a comparison of ordinary Sanchez-Maranon M, Soriano M, Melgosa M, Delgado G, Delgado R ındicator kriging and disjunctive kriging. Geoderma 118:39- (2004). Quantifying the effects of aggregation, particle size and 53.http://dx.doi.org/10.1016/S0016-7061(03)00168-X components on the colour of Mediterranean soils Eur. J. Soil Sci. McBratney AB, Pringle MJ (1999). Estimating average and proportional 55:551-565.http://dx.doi.org/10.1111/j.1365-2389.2004.00624.x variograms of soil properties and their potential use in precision Scheinost AC, Schwertmann U (1999). Color identification of iron oxides agriculture. Precision Agric. 1:125- and hydroxysulfates: use and limitations. Soil Sci. Soc. Am. J. 152.http://dx.doi.org/10.1023/A:1009995404447 63:1463-1471. Melville MD, Atkinson G (1985). Soil color: its measurement and its Schloeder CA, Zimmerman NE, Jacobs MJ (2001). Comparison of designation in models of uniform color space. J. Soil Sci. 36:495- methods for ınterpolating soil properties using limited data. Soil Sci. 512.http://dx.doi.org/10.1111/j.1365-2389.1985.tb00353.x Soc. Am. J. 65:470-479.http://dx.doi.org/10.2136/sssaj2001.652470x Munsell Color Company (1994). Munsell Soil color Charts, Revised Soil Survey Staff (1999). Soil taxonomy: A basic system of soil Edition,Macbeth Division of Kollmorgen, New Windsor, NY. classification for making and interpreting soil survey. U.S. Govt. Nelson RE (1982). Carbonate and gypsum. Page et al. (ed.) Methods of Printing Office, Washington, DC. Agric. Handbook, P. 436. soil analysis. Part II. 2nd ed. Agronomy Monograph. 9. ASA and Spielvogel S, Knicker H, Kögel-Knabner I (2004). SSSA, Madison, WI, pp. 181–197. composition and soil lightness. J. Plant Nutr. Soil Sci. 167:545- Nelson DW, Sommers LE (1982). Total carbon, organic carbon, and 555.http://dx.doi.org/10.1002/jpln.200421424 organic matter. Methods of soil analysis, Part 2. Chemical and SPSS (2000). SPSS for Windows. Student Version. Release 10.0.9 microbiological properties, 2nd edition. SSSA Inc. Publisher, SPSS Inc. IL USA Madison, Wisconsin, pp. 961-1010. Torrent J, Barron V (1993). Laboratory measurement of soil color: Olsen SR, Cole, CV, Watanabe, FS, Dean LA (1954). Estimation of theory and practice. In: Bigham, JM, Ciolkosz, EJ (Eds.), Soil Color. available phosphorus in soils by extraction with sodium bicarbonate Soil Sci. Soc. Am. Madison, WI, pp. 21-34. US Dept. Agriculture Cric, pp. 939-1954. Viscarra Rossel RA, Minasny B, Roudier P, McBratney AB (2006). Ozgoz E, Akbas F, Gunal H, Ersahin S, Çetin M (2009). Long term Colour space model for soil science. Geoderma 133(3-4):320- conventional tillage effect on spatial variability of some soil physical 337.http://dx.doi.org/10.1016/j.geoderma.2005.07.017 properties. J. Sustain. Agric. 33:(2) 142-160. Erratum 1: 33(4):487 Webster R, Oliver MA (2007). Geostatistics for Environmentals. John Erratum 2: 33(8):940. Wiley & Sons Ltd. 2nd Edition, P. 271. Rochette P, Desjardins RL, Pattey E (1991). Spatial and temporal variability of in agricultural fields. Can. J Soil Sci. 71:189–196.http://dx.doi.org/10.4141/cjss91-018 Rossi RE, Mulla DJ, Franz EH (1992). Geostatistical tools for modeling and interpreting ecological spatial dependence. Ecological